
#Data Interpretation
import pandas as pd
import numpy as np
import missingno as msno
import scipy.stats as st
import cufflinks as cf

#Data Visualization
import matplotlib.pyplot as plt
import seaborn as sns
from mpl_toolkits.mplot3d import Axes3D

#Data Statistics
from IPython.display import display_html
from sklearn.preprocessing import StandardScaler
from sklearn import metrics
from prophet import Prophet


data = pd.read_csv("stations.csv")


data


stations = data.dropna()


stations


stations.reset_index(inplace = True)


station_city_state = stations[['StationId','StationName','City','State']]


station_city_state


station_day = pd.read_csv("station_day.csv")


station_day


station_merged_day = pd.merge(station_city_state , station_day, on = 'StationId', how = 'inner')


station_merged_day


#For putting some upper and lower values in the empty space from the required data.
stations_update_day = station_merged_day.fillna(method="bfill", limit=1)


#Updating Date and column names.
stations_update_day['Date'] = pd.to_datetime(stations_update_day['Date'])
stations_update_day.rename(columns = {'AQI_Bucket':'Air_quality'}, inplace = True)


stations_update_day


city_day = pd.read_csv("city_day.csv")


city_day


#For putting some upper and lower values in the empty space from the required data.
city_day_update = city_day.fillna(method="bfill", limit=1)


#Updating Date and column names.
city_day_update['Date'] = pd.to_datetime(city_day_update['Date'])
city_day_update.rename(columns = {'AQI_Bucket':'Air_quality'}, inplace = True)


city_day_update


print(stations_update_day.shape)

(107711, 19)


stations_update_day.isnull().sum()

StationId          0
StationName        0
City               0
State              0
Date               0
PM2.5          20417
PM10           41789
NO             15629
NO2            15058
NOx            14346
NH3            47245
CO             11386
SO2            23922
O3             24213
Benzene        30164
Toluene        37453
Xylene         84595
AQI            18958
Air_quality    18958
dtype: int64


msno.heatmap(stations_update_day)

<AxesSubplot:>


# Calculating Missing Values(Function-1).
def missing_values_table(df):
        # Total missing values
        mis_value = df.isnull().sum()
        
        # Percentage of missing values
        mis_value_percent = 100 * df.isnull().sum() / len(df)
        
        # Make a table with the results
        mis_value_table = pd.concat([mis_value, mis_value_percent], axis=1)
        
        # Rename the columns
        mis_value_table_ren_columns = mis_value_table.rename(
        columns = {0 : 'Missing Values', 1 : '% of Total Values'})
        
        # Sort the table by percentage of missing descending
        mis_value_table_ren_columns = mis_value_table_ren_columns[
            mis_value_table_ren_columns.iloc[:,1] != 0].sort_values(
        '% of Total Values', ascending=False).round(1)
        
        # Print some summary information
        print ("Your selected dataframe has " + str(df.shape[1]) + " columns.\n"      
            "There are " + str(mis_value_table_ren_columns.shape[0]) +
              " columns that have missing values.")
        
        # Return the dataframe with missing information
        return mis_value_table_ren_columns

missing_values= missing_values_table(stations_update_day)
missing_values.style.background_gradient(cmap='Reds')

Your selected dataframe has 19 columns.
There are 14 columns that have missing values.


stations_update_day.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 107711 entries, 0 to 107710
Data columns (total 19 columns):
 #   Column       Non-Null Count   Dtype         
---  ------       --------------   -----         
 0   StationId    107711 non-null  object        
 1   StationName  107711 non-null  object        
 2   City         107711 non-null  object        
 3   State        107711 non-null  object        
 4   Date         107711 non-null  datetime64[ns]
 5   PM2.5        87294 non-null   float64       
 6   PM10         65922 non-null   float64       
 7   NO           92082 non-null   float64       
 8   NO2          92653 non-null   float64       
 9   NOx          93365 non-null   float64       
 10  NH3          60466 non-null   float64       
 11  CO           96325 non-null   float64       
 12  SO2          83789 non-null   float64       
 13  O3           83498 non-null   float64       
 14  Benzene      77547 non-null   float64       
 15  Toluene      70258 non-null   float64       
 16  Xylene       23116 non-null   float64       
 17  AQI          88753 non-null   float64       
 18  Air_quality  88753 non-null   object        
dtypes: datetime64[ns](1), float64(13), object(5)
memory usage: 16.4+ MB


for i in stations_update_day.columns:
    print('column name:{}   unique values:{}'.format(i,len(stations_update_day[i].unique())))

column name:StationId   unique values:108
column name:StationName   unique values:108
column name:City   unique values:24
column name:State   unique values:21
column name:Date   unique values:2009
column name:PM2.5   unique values:22392
column name:PM10   unique values:29547
column name:NO   unique values:11914
column name:NO2   unique values:12051
column name:NOx   unique values:15585
column name:NH3   unique values:9112
column name:CO   unique values:2353
column name:SO2   unique values:5802
column name:O3   unique values:11161
column name:Benzene   unique values:3018
column name:Toluene   unique values:8714
column name:Xylene   unique values:1893
column name:AQI   unique values:931
column name:Air_quality   unique values:7


stations_update_day1 = stations_update_day.copy()


stations_update_day1['BTX'] = stations_update_day1['Benzene']+stations_update_day1['Toluene']+stations_update_day1['Xylene']
stations_update_day1.drop(['Benzene','Toluene','Xylene'],axis=1,inplace = True);


stations_update_day1['Particulate_Matter'] = stations_update_day1['PM2.5']+stations_update_day1['PM10']


stations_update_day1


pollutants_stations = ['PM2.5','PM10','NO','NO2','NOx','NH3','CO','SO2','O3','BTX','Particulate_Matter']


stations_update_day1.set_index('Date',inplace=True)
axes = stations_update_day1[pollutants_stations].plot(marker='.', alpha=0.5, linestyle='None', figsize=(26, 30), subplots=True)
for ax in axes:
    
    ax.set_xlabel('Years')
    ax.set_ylabel('ug / m3')


print(city_day_update.shape)

(29531, 16)


city_day_update.isnull().sum()

City               0
Date               0
PM2.5           4321
PM10           10866
NO              3276
NO2             3278
NOx             3980
NH3            10061
CO              1745
SO2             3510
O3              3664
Benzene         5298
Toluene         7739
Xylene         17878
AQI             4174
Air_quality     4174
dtype: int64


msno.heatmap(city_day_update)

<AxesSubplot:>


# Calculating Missing Values(by help of Function - 1 created above).
missing_values_city = missing_values_table(city_day_update)
missing_values_city.style.background_gradient(cmap='Oranges')

Your selected dataframe has 16 columns.
There are 14 columns that have missing values.


city_day_update.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 29531 entries, 0 to 29530
Data columns (total 16 columns):
 #   Column       Non-Null Count  Dtype         
---  ------       --------------  -----         
 0   City         29531 non-null  object        
 1   Date         29531 non-null  datetime64[ns]
 2   PM2.5        25210 non-null  float64       
 3   PM10         18665 non-null  float64       
 4   NO           26255 non-null  float64       
 5   NO2          26253 non-null  float64       
 6   NOx          25551 non-null  float64       
 7   NH3          19470 non-null  float64       
 8   CO           27786 non-null  float64       
 9   SO2          26021 non-null  float64       
 10  O3           25867 non-null  float64       
 11  Benzene      24233 non-null  float64       
 12  Toluene      21792 non-null  float64       
 13  Xylene       11653 non-null  float64       
 14  AQI          25357 non-null  float64       
 15  Air_quality  25357 non-null  object        
dtypes: datetime64[ns](1), float64(13), object(2)
memory usage: 3.6+ MB


for i in city_day_update.columns:
    print('column name:{}   unique values:{}'.format(i,len(city_day_update[i].unique())))

column name:City   unique values:26
column name:Date   unique values:2009
column name:PM2.5   unique values:11717
column name:PM10   unique values:12572
column name:NO   unique values:5777
column name:NO2   unique values:7405
column name:NOx   unique values:8157
column name:NH3   unique values:5923
column name:CO   unique values:1780
column name:SO2   unique values:4762
column name:O3   unique values:7700
column name:Benzene   unique values:1874
column name:Toluene   unique values:3609
column name:Xylene   unique values:1562
column name:AQI   unique values:830
column name:Air_quality   unique values:7


city_day_update1 = city_day_update.copy()


city_day_update1['BTX'] = city_day_update1['Benzene']+city_day_update1['Toluene']+city_day_update1['Xylene']
city_day_update1.drop(['Benzene','Toluene','Xylene'],axis=1,inplace = True)


city_day_update1['Particulate_Matter'] = city_day_update1['PM2.5']+city_day_update1['PM10']


city_day_update1


pollutants_city = ['PM2.5','PM10','NO','NO2','NOx','NH3','CO','SO2','O3','BTX','Particulate_Matter']


city_day_update1.set_index('Date',inplace=True)
axes = city_day_update1[pollutants_city].plot(marker='.', alpha=0.5, linestyle='None', figsize=(26, 30), subplots=True)
for ax in axes:
    
    ax.set_xlabel('Years')
    ax.set_ylabel('ug / m3')


stations_update_day1


#Function for printing Most polluted stations.
def max_polluted_stn(pollutants_stations):
    x1 = stations_update_day1[[pollutants_stations,'StationName','City']].groupby(["StationName"]).mean().sort_values(by=pollutants_stations,ascending=False).reset_index()
    x1[pollutants_stations] = round(x1[pollutants_stations],2)
    return x1[:10].style.background_gradient(cmap='OrRd')


#For printing the table side by side.
#Source - https://stackoverflow.com/questions/38783027/jupyter-notebook-display-two-pandas-tables-side-by-side
def display_side_by_side(*args):
    html_str=''
    for df in args:
        html_str+=df.render()
    display_html(html_str.replace('table','table style="display:inline"'),raw=True)


pm2_5 = max_polluted_stn('PM2.5')
pm10 = max_polluted_stn('PM10')
Particulate_Matter = max_polluted_stn('Particulate_Matter')
no = max_polluted_stn('NO')
no2 = max_polluted_stn('NO2')
nox = max_polluted_stn('NOx')
nh3 = max_polluted_stn('NH3')
co = max_polluted_stn('CO')
so2 = max_polluted_stn('SO2')
o3 = max_polluted_stn('O3')
btx = max_polluted_stn('BTX')
display_side_by_side(pm2_5,pm10,Particulate_Matter,no,no2,nox,nh3,co,so2,o3,btx)


#Function for printing Least polluted stations.
def min_polluted_stn(pollutants_stations):
    x1 = stations_update_day1[[pollutants_stations,'StationName','City']].groupby(["StationName"]).mean().sort_values(by=pollutants_stations,ascending=True).reset_index()
    x1[pollutants_stations] = round(x1[pollutants_stations],2)
    return x1[:10].style.background_gradient(cmap='GnBu')


#For printing the table side by side.
#Source - https://stackoverflow.com/questions/38783027/jupyter-notebook-display-two-pandas-tables-side-by-side
def display_side_by_side(*args):
    html_str=''
    for df in args:
        html_str+=df.render()
    display_html(html_str.replace('table','table style="display:inline"'),raw=True)


pm2_5 = min_polluted_stn('PM2.5')
pm10 = min_polluted_stn('PM10')
Particulate_Matter = min_polluted_stn('Particulate_Matter')
no = min_polluted_stn('NO')
no2 = min_polluted_stn('NO2')
nox = min_polluted_stn('NOx')
nh3 = min_polluted_stn('NH3')
co = min_polluted_stn('CO')
so2 = min_polluted_stn('SO2')
o3 = min_polluted_stn('O3')
btx = min_polluted_stn('BTX')
display_side_by_side(pm2_5,pm10,Particulate_Matter,no,no2,nox,nh3,co,so2,o3,btx)


pollutants_city2 = ['PM2.5','PM10','NO','NO2','NOx','NH3','CO','SO2','O3','BTX']


#Function for printing Most polluted stations.
def max_polluted_city(pollutants_city2):
    x1 = city_day_update1[[pollutants_city2,'City']].groupby(["City"]).mean().sort_values(by=pollutants_city2,ascending=False).reset_index()
    x1[pollutants_city2] = round(x1[pollutants_city2],2)
    return x1[:10].style.background_gradient(cmap='OrRd')


#For printing the table side by side.
#Source - https://stackoverflow.com/questions/38783027/jupyter-notebook-display-two-pandas-tables-side-by-side
def display_side_by_side(*args):
    html_str=''
    for df in args:
        html_str+=df.render()
    display_html(html_str.replace('table','table style="display:inline"'),raw=True)


pm2_5 = max_polluted_city('PM2.5')
pm10 = max_polluted_city('PM10')
no = max_polluted_city('NO')
no2 = max_polluted_city('NO2')
nox = max_polluted_city('NOx')
nh3 = max_polluted_city('NH3')
co = max_polluted_city('CO')
so2 = max_polluted_city('SO2')
o3 = max_polluted_city('O3')
btx = max_polluted_city('BTX')
display_side_by_side(pm2_5,pm10,no,no2,nox,nh3,co,so2,o3,btx)


#Function for printing Least polluted stations.
def min_polluted_city(pollutants_city2):
    x1 = city_day_update1[[pollutants_city2,'City']].groupby(["City"]).mean().sort_values(by=pollutants_city2,ascending=True).reset_index()
    x1[pollutants_city2] = round(x1[pollutants_city2],2)
    return x1[:10].style.background_gradient(cmap='GnBu')


#For printing the table side by side.
#Source - https://stackoverflow.com/questions/38783027/jupyter-notebook-display-two-pandas-tables-side-by-side
def display_side_by_side(*args):
    html_str=''
    for df in args:
        html_str+=df.render()
    display_html(html_str.replace('table','table style="display:inline"'),raw=True)


pm2_5 = min_polluted_city('PM2.5')
pm10 = min_polluted_city('PM10')
no = min_polluted_city('NO')
no2 = min_polluted_city('NO2')
nox = min_polluted_city('NOx')
nh3 = min_polluted_city('NH3')
co = min_polluted_city('CO')
so2 = min_polluted_city('SO2')
o3 = min_polluted_city('O3')
btx = min_polluted_city('BTX')
display_side_by_side(pm2_5,pm10,no,no2,nox,nh3,co,so2,o3,btx)


stations_update_day


stations_update_day.isnull().sum()

StationId          0
StationName        0
City               0
State              0
Date               0
PM2.5          20417
PM10           41789
NO             15629
NO2            15058
NOx            14346
NH3            47245
CO             11386
SO2            23922
O3             24213
Benzene        30164
Toluene        37453
Xylene         84595
AQI            18958
Air_quality    18958
dtype: int64


stations_update_day.describe()


#Replacing nullspaces with median values.
stations_update_day['PM2.5']=stations_update_day['PM2.5'].fillna(stations_update_day['PM2.5'].median())

stations_update_day['PM10']=stations_update_day['PM10'].fillna(stations_update_day['PM10'].median())

stations_update_day['NO']=stations_update_day['NO'].fillna(stations_update_day['NO'].median())

stations_update_day['NO2']=stations_update_day['NO2'].fillna(stations_update_day['NO2'].median())

stations_update_day['NOx']=stations_update_day['NOx'].fillna(stations_update_day['NOx'].median())

stations_update_day['NH3']=stations_update_day['NH3'].fillna(stations_update_day['NH3'].median())

stations_update_day['CO']=stations_update_day['CO'].fillna(stations_update_day['CO'].median())

stations_update_day['SO2']=stations_update_day['SO2'].fillna(stations_update_day['SO2'].median())

stations_update_day['O3']=stations_update_day['O3'].fillna(stations_update_day['O3'].median())

stations_update_day['Benzene']=stations_update_day['Benzene'].fillna(stations_update_day['Benzene'].median())

stations_update_day['Toluene']=stations_update_day['Toluene'].fillna(stations_update_day['Toluene'].median())

stations_update_day['Xylene']=stations_update_day['Xylene'].fillna(stations_update_day['Xylene'].median())

stations_update_day['AQI']=stations_update_day['AQI'].fillna(stations_update_day['AQI'].median())

stations_update_day['Air_quality']=stations_update_day['Air_quality'].fillna('Moderate')


#Counting Number of zeroes.
(stations_update_day== 0).sum(axis=0)

StationId          0
StationName        0
City               0
State              0
Date               0
PM2.5              0
PM10               0
NO                 0
NO2                0
NOx             4878
NH3                0
CO              7484
SO2                0
O3                 0
Benzene        12876
Toluene        10550
Xylene          6146
AQI                0
Air_quality        0
dtype: int64


#Replacing zeroes with medians.
stations_update_day['PM2.5']=stations_update_day['PM2.5'].replace(0,stations_update_day['PM2.5'].median())

stations_update_day['PM10']=stations_update_day['PM10'].replace(0,stations_update_day['PM10'].median())

stations_update_day['NO']=stations_update_day['NO'].replace(0,stations_update_day['NO'].median())

stations_update_day['NO2']=stations_update_day['NO2'].replace(0,stations_update_day['NO2'].median())

stations_update_day['NOx']=stations_update_day['NOx'].replace(0,stations_update_day['NOx'].median())

stations_update_day['NH3']=stations_update_day['NH3'].replace(0,stations_update_day['NH3'].median())

stations_update_day['CO']=stations_update_day['CO'].replace(0,stations_update_day['CO'].median())

stations_update_day['SO2']=stations_update_day['SO2'].replace(0,stations_update_day['SO2'].median())

stations_update_day['O3']=stations_update_day['O3'].replace(0,stations_update_day['O3'].median())

stations_update_day['Benzene']=stations_update_day['Benzene'].replace(0,stations_update_day['Benzene'].median())

stations_update_day['Toluene']=stations_update_day['Toluene'].replace(0,stations_update_day['Toluene'].median())

stations_update_day['Xylene']=stations_update_day['Xylene'].replace(0,stations_update_day['Xylene'].median())


(stations_update_day== 0).sum(axis=0)

StationId      0
StationName    0
City           0
State          0
Date           0
PM2.5          0
PM10           0
NO             0
NO2            0
NOx            0
NH3            0
CO             0
SO2            0
O3             0
Benzene        0
Toluene        0
Xylene         0
AQI            0
Air_quality    0
dtype: int64


stations_update_day.isnull().sum()

StationId      0
StationName    0
City           0
State          0
Date           0
PM2.5          0
PM10           0
NO             0
NO2            0
NOx            0
NH3            0
CO             0
SO2            0
O3             0
Benzene        0
Toluene        0
Xylene         0
AQI            0
Air_quality    0
dtype: int64


stations_update_day.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 107711 entries, 0 to 107710
Data columns (total 19 columns):
 #   Column       Non-Null Count   Dtype         
---  ------       --------------   -----         
 0   StationId    107711 non-null  object        
 1   StationName  107711 non-null  object        
 2   City         107711 non-null  object        
 3   State        107711 non-null  object        
 4   Date         107711 non-null  datetime64[ns]
 5   PM2.5        107711 non-null  float64       
 6   PM10         107711 non-null  float64       
 7   NO           107711 non-null  float64       
 8   NO2          107711 non-null  float64       
 9   NOx          107711 non-null  float64       
 10  NH3          107711 non-null  float64       
 11  CO           107711 non-null  float64       
 12  SO2          107711 non-null  float64       
 13  O3           107711 non-null  float64       
 14  Benzene      107711 non-null  float64       
 15  Toluene      107711 non-null  float64       
 16  Xylene       107711 non-null  float64       
 17  AQI          107711 non-null  float64       
 18  Air_quality  107711 non-null  object        
dtypes: datetime64[ns](1), float64(13), object(5)
memory usage: 16.4+ MB


stations_update_day.describe()


stations_update_day2 = stations_update_day.copy()


stations_update_day2['StationName'].unique()

array(['Secretariat, Amaravati - APPCB',
       'GVM Corporation, Visakhapatnam - APPCB',
       'Railway Colony, Guwahati - APCB',
       'DRM Office Danapur, Patna - BSPCB',
       'Govt. High School Shikarpur, Patna - BSPCB',
       'IGSC Planetarium Complex, Patna - BSPCB',
       'Muradpur, Patna - BSPCB', 'Rajbansi Nagar, Patna - BSPCB',
       'Samanpura, Patna - BSPCB', 'Sector-25, Chandigarh - CPCC',
       'Alipur, Delhi - DPCC', 'Anand Vihar, Delhi - DPCC',
       'Ashok Vihar, Delhi - DPCC', 'Aya Nagar, Delhi - IMD',
       'Bawana, Delhi - DPCC', 'Burari Crossing, Delhi - IMD',
       'CRRI Mathura Road, Delhi - IMD', 'DTU, Delhi - CPCB',
       'Dr. Karni Singh Shooting Range, Delhi - DPCC',
       'Dwarka-Sector 8, Delhi - DPCC', 'East Arjun Nagar, Delhi - CPCB',
       'IGI Airport (T3), Delhi - IMD',
       'IHBAS, Dilshad Garden, Delhi - CPCB', 'ITO, Delhi - CPCB',
       'Jahangirpuri, Delhi - DPCC',
       'Jawaharlal Nehru Stadium, Delhi - DPCC',
       'Lodhi Road, Delhi - IMD',
       'Major Dhyan Chand National Stadium, Delhi - DPCC',
       'Mandir Marg, Delhi - DPCC', 'Mundka, Delhi - DPCC',
       'NSIT Dwarka, Delhi - CPCB', 'Najafgarh, Delhi - DPCC',
       'Narela, Delhi - DPCC', 'Nehru Nagar, Delhi - DPCC',
       'North Campus, DU, Delhi - IMD', 'Okhla Phase-2, Delhi - DPCC',
       'Patparganj, Delhi - DPCC', 'Punjabi Bagh, Delhi - DPCC',
       'Pusa, Delhi - DPCC', 'Pusa, Delhi - IMD',
       'R K Puram, Delhi - DPCC', 'Rohini, Delhi - DPCC',
       'Shadipur, Delhi - CPCB', 'Sirifort, Delhi - CPCB',
       'Sonia Vihar, Delhi - DPCC', 'Sri Aurobindo Marg, Delhi - DPCC',
       'Vivek Vihar, Delhi - DPCC', 'Wazirpur, Delhi - DPCC',
       'Maninagar, Ahmedabad - GPCB', 'NISE Gwal Pahari, Gurugram - IMD',
       'Sector-51, Gurugram - HSPCB', 'Teri Gram, Gurugram - HSPCB',
       'Vikas Sadan, Gurugram - HSPCB',
       'Tata Stadium, Jorapokhar - JSPCB', 'BTM Layout, Bengaluru - CPCB',
       'BWSSB Kadabesanahalli, Bengaluru - CPCB',
       'Bapuji Nagar, Bengaluru - KSPCB',
       'City Railway Station, Bengaluru - KSPCB',
       'Hebbal, Bengaluru - KSPCB', 'Hombegowda Nagar, Bengaluru - KSPCB',
       'Jayanagar 5th Block, Bengaluru - KSPCB',
       'Peenya, Bengaluru - CPCB', 'Sanegurava Halli, Bengaluru - KSPCB',
       'Silk Board, Bengaluru - KSPCB',
       'Kariavattom, Thiruvananthapuram - Kerala PCB',
       'Plammoodu, Thiruvananthapuram - Kerala PCB',
       'T T Nagar, Bhopal - MPPCB', 'Bandra, Mumbai - MPCB',
       'Borivali East, Mumbai - MPCB',
       'Chhatrapati Shivaji Intl. Airport (T2), Mumbai - MPCB',
       'Colaba, Mumbai - MPCB', 'Kurla, Mumbai - MPCB',
       'Powai, Mumbai - MPCB', 'Sion, Mumbai - MPCB',
       'Vasai West, Mumbai - MPCB', 'Vile Parle West, Mumbai - MPCB',
       'Worli, Mumbai - MPCB', 'Lumpyngngad, Shillong - Meghalaya PCB',
       'Sikulpuikawn, Aizawl - Mizoram PCB',
       'GM Office, Brajrajnagar - OSPCB',
       'Talcher Coalfields,Talcher - OSPCB',
       'Golden Temple, Amritsar - PPCB', 'Adarsh Nagar, Jaipur - RSPCB',
       'Police Commissionerate, Jaipur - RSPCB',
       'Shastri Nagar, Jaipur - RSPCB',
       'Alandur Bus Depot, Chennai - CPCB',
       'Manali Village, Chennai - TNPCB', 'Manali, Chennai - CPCB',
       'Velachery Res. Area, Chennai - CPCB',
       'SIDCO Kurichi, Coimbatore - TNPCB',
       'Bollaram Industrial Area, Hyderabad - TSPCB',
       'Central University, Hyderabad - TSPCB',
       'ICRISAT Patancheru, Hyderabad - TSPCB',
       'IDA Pashamylaram, Hyderabad - TSPCB',
       'Sanathnagar, Hyderabad - TSPCB', 'Zoo Park, Hyderabad - TSPCB',
       'Central School, Lucknow - CPCB', 'Gomti Nagar, Lucknow - UPPCB',
       'Lalbagh, Lucknow - CPCB', 'Nishant Ganj, Lucknow - UPPCB',
       'Talkatora District Industries Center, Lucknow - CPCB',
       'Ballygunge, Kolkata - WBPCB', 'Bidhannagar, Kolkata - WBPCB',
       'Fort William, Kolkata - WBPCB', 'Jadavpur, Kolkata - WBPCB',
       'Rabindra Bharati University, Kolkata - WBPCB',
       'Rabindra Sarobar, Kolkata - WBPCB', 'Victoria, Kolkata - WBPCB'],
      dtype=object)


stations_update_day2['StationName'].value_counts()

IHBAS, Dilshad Garden, Delhi - CPCB           2009
Manali, Chennai - CPCB                        2009
NSIT Dwarka, Delhi - CPCB                     2009
Bandra, Mumbai - MPCB                         2009
Maninagar, Ahmedabad - GPCB                   2009
                                              ... 
DRM Office Danapur, Patna - BSPCB              126
Govt. High School Shikarpur, Patna - BSPCB     121
Teri Gram, Gurugram - HSPCB                    119
Sector-51, Gurugram - HSPCB                    119
Sikulpuikawn, Aizawl - Mizoram PCB             113
Name: StationName, Length: 108, dtype: int64


stations_update_day2['Pollution content'] = stations_update_day2['PM2.5']+stations_update_day2['PM10']+stations_update_day2['NO']+stations_update_day2['NO2']+stations_update_day2['NOx']+stations_update_day2['NH3']+stations_update_day2['CO']+stations_update_day2['SO2']+stations_update_day2['O3']+stations_update_day2['Benzene']+stations_update_day2['Toluene']+stations_update_day2['Xylene']


stations_update_day2


cf.go_offline()
cf.set_config_file(offline=False, world_readable=True)


#Ploting the most polluted stations in Bar form.
def max_bar_plot(var):
    x1 = stations_update_day2[['StationName',var]].groupby(["StationName"]).median().sort_values(by = var,
    ascending = True).tail(10).iplot(kind='bar', xTitle='Stations',yTitle=var, 
                                     linecolor='black', title='{2} {1} {0}'.format(")",var,' Most polluted Stations('))

p = max_bar_plot('Pollution content')


#Ploting the least polluted stations in Bar form.
def min_bar_plot(var):
    x1 = stations_update_day2[['StationName',var]].groupby(["StationName"]).median().sort_values(by = var,
    ascending = True).head(10).iplot(kind='bar', xTitle='Stations',yTitle=var, 
                                     linecolor='black', title='{2} {1} {0}'.format(")",var,' Minimum polluted Stations('))

p1 = min_bar_plot('Pollution content')


stations_update_day.to_csv("stations_update_day.csv", index = False)


city_day_update2 = city_day_update.copy()


city_day_update2.isnull().sum()

City               0
Date               0
PM2.5           4321
PM10           10866
NO              3276
NO2             3278
NOx             3980
NH3            10061
CO              1745
SO2             3510
O3              3664
Benzene         5298
Toluene         7739
Xylene         17878
AQI             4174
Air_quality     4174
dtype: int64


city_day_update2.describe()


#Replacing nullspaces with median values.
city_day_update2['PM2.5']=city_day_update2['PM2.5'].fillna(city_day_update2['PM2.5'].median())

city_day_update2['PM10']=city_day_update2['PM10'].fillna(city_day_update2['PM10'].median())

city_day_update2['NO']=city_day_update2['NO'].fillna(city_day_update2['NO'].median())

city_day_update2['NO2']=city_day_update2['NO2'].fillna(city_day_update2['NO2'].median())

city_day_update2['NOx']=city_day_update2['NOx'].fillna(city_day_update2['NOx'].median())

city_day_update2['NH3']=city_day_update2['NH3'].fillna(city_day_update2['NH3'].median())

city_day_update2['CO']=city_day_update2['CO'].fillna(city_day_update2['CO'].median())

city_day_update2['SO2']=city_day_update2['SO2'].fillna(city_day_update2['SO2'].median())

city_day_update2['O3']=city_day_update2['O3'].fillna(city_day_update2['O3'].median())

city_day_update2['Benzene']=city_day_update2['Benzene'].fillna(city_day_update2['Benzene'].median())

city_day_update2['Toluene']=city_day_update2['Toluene'].fillna(city_day_update2['Toluene'].median())

city_day_update2['Xylene']=city_day_update2['Xylene'].fillna(city_day_update2['Xylene'].median())

city_day_update2['AQI']=city_day_update2['AQI'].fillna(city_day_update2['AQI'].median())

city_day_update2['Air_quality']=city_day_update2['Air_quality'].fillna('Moderate')


#Counting Number of zeroes.
(city_day_update2== 0).sum(axis=0)

City              0
Date              0
PM2.5             0
PM10              0
NO                0
NO2               0
NOx             757
NH3               0
CO             2413
SO2               0
O3                0
Benzene        3913
Toluene        2943
Xylene         1812
AQI               0
Air_quality       0
dtype: int64


#Replacing zeroes with medians.
city_day_update2['PM2.5']=city_day_update2['PM2.5'].replace(0,city_day_update2['PM2.5'].median())

city_day_update2['PM10']=city_day_update2['PM10'].replace(0,city_day_update2['PM10'].median())

city_day_update2['NO']=city_day_update2['NO'].replace(0,city_day_update2['NO'].median())

city_day_update2['NO2']=city_day_update2['NO2'].replace(0,city_day_update2['NO2'].median())

city_day_update2['NOx']=city_day_update2['NOx'].replace(0,city_day_update2['NOx'].median())

city_day_update2['NH3']=city_day_update2['NH3'].replace(0,city_day_update2['NH3'].median())

city_day_update2['CO']=city_day_update2['CO'].replace(0,city_day_update2['CO'].median())

city_day_update2['SO2']=city_day_update2['SO2'].replace(0,city_day_update2['SO2'].median())

city_day_update2['O3']=city_day_update2['O3'].replace(0,city_day_update2['O3'].median())

city_day_update2['Benzene']=city_day_update2['Benzene'].replace(0,city_day_update2['Benzene'].median())

city_day_update2['Toluene']=city_day_update2['Toluene'].replace(0,city_day_update2['Toluene'].median())

city_day_update2['Xylene']=city_day_update2['Xylene'].replace(0,city_day_update2['Xylene'].median())


(city_day_update2 == 0).sum(axis=0)

City           0
Date           0
PM2.5          0
PM10           0
NO             0
NO2            0
NOx            0
NH3            0
CO             0
SO2            0
O3             0
Benzene        0
Toluene        0
Xylene         0
AQI            0
Air_quality    0
dtype: int64


city_day_update2.isnull().sum()

City           0
Date           0
PM2.5          0
PM10           0
NO             0
NO2            0
NOx            0
NH3            0
CO             0
SO2            0
O3             0
Benzene        0
Toluene        0
Xylene         0
AQI            0
Air_quality    0
dtype: int64


city_day_update2.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 29531 entries, 0 to 29530
Data columns (total 16 columns):
 #   Column       Non-Null Count  Dtype         
---  ------       --------------  -----         
 0   City         29531 non-null  object        
 1   Date         29531 non-null  datetime64[ns]
 2   PM2.5        29531 non-null  float64       
 3   PM10         29531 non-null  float64       
 4   NO           29531 non-null  float64       
 5   NO2          29531 non-null  float64       
 6   NOx          29531 non-null  float64       
 7   NH3          29531 non-null  float64       
 8   CO           29531 non-null  float64       
 9   SO2          29531 non-null  float64       
 10  O3           29531 non-null  float64       
 11  Benzene      29531 non-null  float64       
 12  Toluene      29531 non-null  float64       
 13  Xylene       29531 non-null  float64       
 14  AQI          29531 non-null  float64       
 15  Air_quality  29531 non-null  object        
dtypes: datetime64[ns](1), float64(13), object(2)
memory usage: 3.6+ MB


city_day_update2.describe()


city_day_bcovid = city_day_update2.copy()


city_day_bcovid = city_day_bcovid[city_day_bcovid['Date'] <= ('01-01-2020')]


city_day_bcovid['Vehicular Pollutants'] = city_day_bcovid['PM2.5']+city_day_bcovid['PM10']+city_day_bcovid['NO']+city_day_bcovid['NO2']+city_day_bcovid['NOx']+city_day_bcovid['NH3']+city_day_bcovid['CO']


city_day_bcovid['Industrial Pollutants'] = city_day_bcovid['SO2']+city_day_bcovid['O3']+city_day_bcovid['Benzene']+city_day_bcovid['Toluene']+city_day_bcovid['Xylene']


city_day_bcovid['Total Pollutants'] = city_day_bcovid['PM2.5']+city_day_bcovid['PM10']+city_day_bcovid['NO']+city_day_bcovid['NO2']+city_day_bcovid['NOx']+city_day_bcovid['NH3']+city_day_bcovid['CO']+city_day_bcovid['SO2']+city_day_bcovid['O3']+city_day_bcovid['Benzene']+city_day_bcovid['Toluene']+city_day_bcovid['Xylene']


city_day_bcovid = city_day_bcovid.drop(['PM2.5', 'PM10', 'NO', 'NO2', 'NOx', 'NH3', 'CO', 'SO2',
       'O3','Benzene','Toluene','Xylene'],axis=1)


city_day_bcovid.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 24908 entries, 0 to 29348
Data columns (total 7 columns):
 #   Column                 Non-Null Count  Dtype         
---  ------                 --------------  -----         
 0   City                   24908 non-null  object        
 1   Date                   24908 non-null  datetime64[ns]
 2   AQI                    24908 non-null  float64       
 3   Air_quality            24908 non-null  object        
 4   Vehicular Pollutants   24908 non-null  float64       
 5   Industrial Pollutants  24908 non-null  float64       
 6   Total Pollutants       24908 non-null  float64       
dtypes: datetime64[ns](1), float64(4), object(2)
memory usage: 1.5+ MB


#Function for plotting.
def ploting(var):
    city_day_bcovid[var].iplot(title=var,xTitle='Cities',yTitle=var, linecolor='black', )
    plt.show()
ploting('Vehicular Pollutants')
ploting('Industrial Pollutants')
ploting('Total Pollutants')


def max_bar_plot(var):
    x1 = city_day_bcovid[['City',var]].groupby(["City"]).median().sort_values(by = var,
    ascending = True).tail(10).iplot(kind='bar', xTitle='Cities',yTitle=var, 
                                     linecolor='black', title='{2} {1} {0}'.format(")",var,' Most polluted cities('))

p1 = max_bar_plot('Industrial Pollutants')
p2 = max_bar_plot('Vehicular Pollutants')
p3 = max_bar_plot('Total Pollutants')


def min_bar_plot(var):
    x1 = city_day_bcovid[['City',var]].groupby(["City"]).mean().sort_values(by = var,
    ascending = True).head(10).iplot(kind='bar', yTitle='Cities',xTitle=var, linecolor='black',title='{2} {1} {0}'.format(")",var,' Least polluted cities ('))
p1 = min_bar_plot('Industrial Pollutants')
p2 = min_bar_plot('Vehicular Pollutants')
p3 = min_bar_plot('Total Pollutants')


def al(var):
    cities = [var]
    city_day_bcovid1 = city_day_bcovid[city_day_bcovid['Date'] <= '01-01-2020']
    AQI = city_day_bcovid1[city_day_bcovid1.City.isin(cities)][['City','Air_quality']]
    AQI[var] = AQI['Air_quality']
    k = AQI[var].value_counts()
    m = pd.DataFrame((round((k/sum(k))*100)))
    return m
c1 = al('Delhi')
c2 = al('Ahmedabad')
c3 = al('Bhopal')
c4 = al('Jaipur')
c5 = al('Patna')
# c6 = al('Lucknow')
# c7 = al('Kolkata')
df_row = pd.concat([c1,c2,c3,c4,c5],axis=1)
df_row.iplot(kind='bar', align='center',xTitle='Satisfaction level', yTitle='Percentage of Satisfaction' ,linecolor='black', title='Satisfaction Level of people (Before COVID-19)')


city_day_bcovid


#city_day_bcovid:mean
vehicular_pollutants_mean = city_day_bcovid['Vehicular Pollutants'].mean()
industrial_pollutants_mean = city_day_bcovid['Industrial Pollutants'].mean()
total_pollutants_mean = city_day_bcovid['Total Pollutants'].mean()

#city_day_bcovid:standard deviation
vehicular_pollutants_var = city_day_bcovid['Vehicular Pollutants'].std() **2
industrial_pollutants_var = city_day_bcovid['Industrial Pollutants'].std() **2
total_pollutants_var = city_day_bcovid['Total Pollutants'].std() **2

pollutant_mean_var = {'Pollutants' : ['Vehicular Pollutants', 'Industrial Pollutants', 'Total Pollutants'],
                      'Mean' : [vehicular_pollutants_mean,industrial_pollutants_mean,total_pollutants_mean],
                      'Variance' : [vehicular_pollutants_var,industrial_pollutants_var,total_pollutants_var]}

bcovid_mean_var = pd.DataFrame(pollutant_mean_var,columns = ['Pollutants','Mean','Variance'])

print(bcovid_mean_var)

              Pollutants        Mean      Variance
0   Vehicular Pollutants  284.058727  28761.160744
1  Industrial Pollutants   60.332131   1715.762624
2       Total Pollutants  344.390859  35606.345506


city_day_bcovid_req = city_day_bcovid[['Vehicular Pollutants','Industrial Pollutants', 'Total Pollutants']]


city_day_bcovid_req


scaler = StandardScaler()
x_scaled = scaler.fit_transform(city_day_bcovid_req)
city_day_bcovid_norm = pd.DataFrame(x_scaled)
city_day_bcovid_norm.rename(columns = {0 : 'Vehicular Pollutants', 1 : 'Industrial Pollutants', 2 : 'Total Pollutants'}, inplace = True)

city_day_bcovid_norm.head()


city_day_bcovid_norm.mean()
#Here mean is apporximated to 0.

Vehicular Pollutants     2.464706e-16
Industrial Pollutants   -2.707183e-16
Total Pollutants        -3.765523e-17
dtype: float64


city_day_bcovid_norm.std()
#Here standard deviation is approximated to 1.

Vehicular Pollutants     1.00002
Industrial Pollutants    1.00002
Total Pollutants         1.00002
dtype: float64


city_day_bcovid_req.skew()

Vehicular Pollutants     2.354917
Industrial Pollutants    2.941972
Total Pollutants         2.157560
dtype: float64


city_day_bcovid_norm.skew()

Vehicular Pollutants     2.354917
Industrial Pollutants    2.941972
Total Pollutants         2.157560
dtype: float64


city_day_bcovid_req.describe()


city_day_bcovid_req.hist(stacked=True, bins=20, figsize=(15,10))

array([[<AxesSubplot:title={'center':'Vehicular Pollutants'}>,
        <AxesSubplot:title={'center':'Industrial Pollutants'}>],
       [<AxesSubplot:title={'center':'Total Pollutants'}>,
        <AxesSubplot:>]], dtype=object)


city_day_bcovid_norm.hist(stacked=True, bins=20, figsize=(15,10))

array([[<AxesSubplot:title={'center':'Vehicular Pollutants'}>,
        <AxesSubplot:title={'center':'Industrial Pollutants'}>],
       [<AxesSubplot:title={'center':'Total Pollutants'}>,
        <AxesSubplot:>]], dtype=object)


city_day_bcovid_req


total_pollutants_mean = city_day_bcovid_req.mean()['Total Pollutants']


total_pollutants_std = city_day_bcovid_req.std()['Total Pollutants']


total_pollutants_zscore = (total_pollutants_mean - 320)/total_pollutants_std


total_pollutants_zscore

0.12925977347574807


total_pollutants_pvalue = st.norm.sf(abs(total_pollutants_zscore))


total_pollutants_pvalue

0.4485760502293482


city_day_afcovid = city_day_update2.copy()


city_day_afcovid = city_day_afcovid[city_day_afcovid['Date'] > ('01-01-2020')]


city_day_afcovid['Vehicular Pollutants'] = city_day_afcovid['PM2.5']+city_day_afcovid['PM10']+city_day_afcovid['NO']+city_day_afcovid['NO2']+city_day_afcovid['NOx']+city_day_afcovid['NH3']+city_day_afcovid['CO']


city_day_afcovid['Industrial Pollutants'] = city_day_afcovid['SO2']+city_day_afcovid['O3']+city_day_afcovid['Benzene']+city_day_afcovid['Toluene']+city_day_afcovid['Xylene']


city_day_afcovid['Total Pollutants'] = city_day_afcovid['PM2.5']+city_day_afcovid['PM10']+city_day_afcovid['NO']+city_day_afcovid['NO2']+city_day_afcovid['NOx']+city_day_afcovid['NH3']+city_day_afcovid['CO']+city_day_afcovid['SO2']+city_day_afcovid['O3']+city_day_afcovid['Benzene']+city_day_afcovid['Toluene']+city_day_afcovid['Xylene']


city_day_afcovid = city_day_afcovid.drop(['PM2.5', 'PM10', 'NO', 'NO2', 'NOx', 'NH3', 'CO', 'SO2',
       'O3','Benzene','Toluene','Xylene'],axis=1)


city_day_afcovid.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 4623 entries, 1827 to 29530
Data columns (total 7 columns):
 #   Column                 Non-Null Count  Dtype         
---  ------                 --------------  -----         
 0   City                   4623 non-null   object        
 1   Date                   4623 non-null   datetime64[ns]
 2   AQI                    4623 non-null   float64       
 3   Air_quality            4623 non-null   object        
 4   Vehicular Pollutants   4623 non-null   float64       
 5   Industrial Pollutants  4623 non-null   float64       
 6   Total Pollutants       4623 non-null   float64       
dtypes: datetime64[ns](1), float64(4), object(2)
memory usage: 288.9+ KB


#Function for plotting.
def ploting(var):
    city_day_afcovid[var].iplot(title=var,xTitle='Cities',yTitle=var, linecolor='black', )
    plt.show()
ploting('Vehicular Pollutants')
ploting('Industrial Pollutants')
ploting('Total Pollutants')


def max_bar_plot(var):
    x1 = city_day_afcovid[['City',var]].groupby(["City"]).median().sort_values(by = var,
    ascending = True).tail(10).iplot(kind='bar', xTitle='Cities',yTitle=var, 
                                     linecolor='black', title='{2} {1} {0}'.format(")",var,' Most polluted cities('))

p1 = max_bar_plot('Industrial Pollutants')
p2 = max_bar_plot('Vehicular Pollutants')
p3 = max_bar_plot('Total Pollutants')


def min_bar_plot(var):
    x1 = city_day_afcovid[['City',var]].groupby(["City"]).mean().sort_values(by = var,
    ascending = True).head(10).iplot(kind='bar', yTitle='Cities',xTitle=var, linecolor='black',title='{2} {1} {0}'.format(")",var,' Least polluted cities ('))
p1 = min_bar_plot('Industrial Pollutants')
p2 = min_bar_plot('Vehicular Pollutants')
p3 = min_bar_plot('Total Pollutants')


def al(var):
    cities = [var]
    city_day_afcovid1 = city_day_afcovid[city_day_afcovid['Date'] > '01-01-2020']
    AQI = city_day_afcovid1[city_day_afcovid1.City.isin(cities)][['City','Air_quality']]
    AQI[var] = AQI['Air_quality']
    k = AQI[var].value_counts()
    m = pd.DataFrame((round((k/sum(k))*100)))
    return m
c1 = al('Delhi')
c2 = al('Ahmedabad')
c3 = al('Bhopal')
c4 = al('Jaipur')
c5 = al('Patna')
c6 = al('Lucknow')
c7 = al('Kolkata')
df_row = pd.concat([c1,c2,c3,c4,c5],axis=1)
df_row.iplot(kind='bar', align='center',xTitle='Satisfaction level', yTitle='Percentage of Satisfaction' ,linecolor='black', title='Satisfaction level of people (After COVID-19)')


city_day_clean = city_day_update2.to_csv("city_day_update.csv", index = False)


city_day_afcovid


#city_day_afcovid:mean
vehicular_pollutants_meanaf = city_day_afcovid['Vehicular Pollutants'].mean()
industrial_pollutants_meanaf = city_day_afcovid['Industrial Pollutants'].mean()
total_pollutants_meanaf = city_day_afcovid['Total Pollutants'].mean()

#city_day_afcovid:standard deviation
vehicular_pollutants_varaf = city_day_afcovid['Vehicular Pollutants'].std() **2
industrial_pollutants_varaf = city_day_afcovid['Industrial Pollutants'].std() **2
total_pollutants_varaf = city_day_afcovid['Total Pollutants'].std() **2

pollutant_mean_varaf = {'Pollutants' : ['Vehicular Pollutants', 'Industrial Pollutants', 'Total Pollutants'],
                      'Mean' : [vehicular_pollutants_meanaf,industrial_pollutants_meanaf,total_pollutants_meanaf],
                      'Variance' : [vehicular_pollutants_varaf,industrial_pollutants_varaf,total_pollutants_varaf]}

afcovid_mean_var = pd.DataFrame(pollutant_mean_varaf,columns = ['Pollutants','Mean','Variance'])

print(afcovid_mean_var)

              Pollutants        Mean      Variance
0   Vehicular Pollutants  219.726823  18957.688067
1  Industrial Pollutants   61.207863   4191.823405
2       Total Pollutants  280.934686  23610.013374


city_day_afcovid_req = city_day_afcovid[['Vehicular Pollutants','Industrial Pollutants', 'Total Pollutants']]


city_day_afcovid_req


scaleraf = StandardScaler()
x_scaledaf = scaleraf.fit_transform(city_day_afcovid_req)
city_day_afcovid_norm = pd.DataFrame(x_scaledaf)
city_day_afcovid_norm.rename(columns = {0 : 'Vehicular Pollutants', 1 : 'Industrial Pollutants', 2 : 'Total Pollutants'}, inplace = True)

city_day_afcovid_norm.head()


city_day_afcovid_norm.mean()
#Here mean is apporximated to 0.

Vehicular Pollutants     1.229579e-17
Industrial Pollutants   -4.303525e-17
Total Pollutants         6.762682e-17
dtype: float64


city_day_afcovid_norm.std()
#Here standard deviation is approximated to 1.

Vehicular Pollutants     1.000108
Industrial Pollutants    1.000108
Total Pollutants         1.000108
dtype: float64


city_day_afcovid_req.skew()

Vehicular Pollutants      1.872608
Industrial Pollutants    10.533733
Total Pollutants          1.835976
dtype: float64


city_day_afcovid_norm.skew()

Vehicular Pollutants      1.872608
Industrial Pollutants    10.533733
Total Pollutants          1.835976
dtype: float64


city_day_afcovid_req.describe()


city_day_afcovid_req.hist(stacked=True, bins=20, figsize=(15,10))

array([[<AxesSubplot:title={'center':'Vehicular Pollutants'}>,
        <AxesSubplot:title={'center':'Industrial Pollutants'}>],
       [<AxesSubplot:title={'center':'Total Pollutants'}>,
        <AxesSubplot:>]], dtype=object)


city_day_afcovid_norm.hist(stacked=True, bins=20, figsize=(15,10))

array([[<AxesSubplot:title={'center':'Vehicular Pollutants'}>,
        <AxesSubplot:title={'center':'Industrial Pollutants'}>],
       [<AxesSubplot:title={'center':'Total Pollutants'}>,
        <AxesSubplot:>]], dtype=object)


city_day_afcovid_req


total_pollutants_mean_af = city_day_afcovid_req.mean()['Total Pollutants']


total_pollutants_std_af = city_day_afcovid_req.std()['Total Pollutants']


total_pollutants_zscore_af = (total_pollutants_mean_af - 320)/total_pollutants_std_af


total_pollutants_zscore_af

-0.25423960141421537


total_pollutants_pvalue_af = st.norm.sf(abs(total_pollutants_zscore_af))


total_pollutants_pvalue_af

0.39965522897962547


city_day_update_temp= city_day_update2.copy()


city_day_update_temp


city_day_update_temp.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 29531 entries, 0 to 29530
Data columns (total 16 columns):
 #   Column       Non-Null Count  Dtype         
---  ------       --------------  -----         
 0   City         29531 non-null  object        
 1   Date         29531 non-null  datetime64[ns]
 2   PM2.5        29531 non-null  float64       
 3   PM10         29531 non-null  float64       
 4   NO           29531 non-null  float64       
 5   NO2          29531 non-null  float64       
 6   NOx          29531 non-null  float64       
 7   NH3          29531 non-null  float64       
 8   CO           29531 non-null  float64       
 9   SO2          29531 non-null  float64       
 10  O3           29531 non-null  float64       
 11  Benzene      29531 non-null  float64       
 12  Toluene      29531 non-null  float64       
 13  Xylene       29531 non-null  float64       
 14  AQI          29531 non-null  float64       
 15  Air_quality  29531 non-null  object        
dtypes: datetime64[ns](1), float64(13), object(2)
memory usage: 3.6+ MB


#Resampling the data for getting the date view of whole country.
city_day_update_pred = pd.DataFrame(city_day_update_temp.groupby('Date')["AQI"].mean()).reset_index()


city_day_update_pred.rename(columns = {'AQI' : 'AQI_avg'}, inplace = True)


#Splitting the data into month and year wise
city_day_update_pred['Month'] = city_day_update_pred['Date'].dt.month
city_day_update_pred['Year'] = city_day_update_pred['Date'].dt.year
city_day_update_pred = city_day_update_pred[['Date','Month','Year','AQI_avg']]
city_day_update_pred


#Predicting the AQI for taking all years in consideration.
city_day_pred_whole = city_day_update_pred.copy()


city_day_pred_whole


#Applying boxplot analysis
city_day_pred_whole = city_day_pred_whole[['AQI_avg','Year']].groupby(["Year"]).mean().reset_index().sort_values(by='Year',ascending=False)
city_day_pred_whole=city_day_pred_whole.dropna()
city_day_pred_whole.describe()

sns.boxplot(x=city_day_pred_whole['AQI_avg'])

<AxesSubplot:xlabel='AQI_avg'>


#Visualizing the filtered data

Year=city_day_pred_whole['Year'].values
AQI=city_day_pred_whole['AQI_avg'].values
city_day_pred_whole['AQI_avg']=pd.to_numeric(city_day_pred_whole['AQI_avg'],errors='coerce')
city_day_pred_whole['Year']=pd.to_numeric(city_day_pred_whole['Year'],errors='coerce')

#Plotting by using 3d graph
plt.rcParams['figure.figsize'] = (20.0, 10.0)
fig = plt.figure()
ax = fig.add_subplot(111, projection="3d")
ax.set_position([0, 0, 0.95, 1])
ax.scatter(Year,AQI, color='red')
plt.show()


#scatter plot of data points
cols =['Year']
y_whole = city_day_pred_whole['AQI_avg']
x_whole = city_day_pred_whole[cols]

plt.scatter(x_whole,y_whole)
plt.show()


x_whole = (x_whole - x_whole.mean()) / x_whole.std()
x_whole = np.c_[np.ones(x_whole.shape[0]), x_whole]
x_whole

array([[ 1.        ,  1.33630621],
       [ 1.        ,  0.80178373],
       [ 1.        ,  0.26726124],
       [ 1.        , -0.26726124],
       [ 1.        , -0.80178373],
       [ 1.        , -1.33630621]])


# Applying GRADIENT DESCENT 

alpha = 0.1 #Step size
iterations = 3000 #No. of iterations
m = y_whole.size #No. of data points
np.random.seed(4) #Setting the seed
theta = np.random.rand(2) #Picking random values to start with

def gradient_descent(x_whole, y_whole, theta, iterations, alpha):
    past_costs = []
    past_thetas = [theta]
    for i in range(iterations):
        prediction = np.dot(x_whole, theta)
        error = prediction - y_whole
        cost = 1/(2*m) * np.dot(error.T, error)
        past_costs.append(cost)
        theta = theta - (alpha * (1/m) * np.dot(x_whole.T, error))
        past_thetas.append(theta)
        
    return past_thetas, past_costs
past_thetas, past_costs = gradient_descent(x_whole, y_whole, theta, iterations, alpha)
theta = past_thetas[-1]

#Printing the results...
print("Gradient Descent: {:.2f}, {:.2f}".format(theta[0], theta[1]))

Gradient Descent: 161.82, -20.54


#Plotting the cost function...
plt.title('Cost Function J')
plt.xlabel('No. of iterations')
plt.ylabel('Cost')
plt.plot(past_costs)
plt.show()


#Predicted value
newB=[ 161.82, -20.54]
def rmse(y_whole,y_pred_whole):
    rmse=np.sqrt(sum(y_whole-y_pred_whole))
    return rmse

   
y_pred_whole=x_whole.dot(newB)

dt = pd.DataFrame({'Actual': y_whole, 'Predicted': y_pred_whole})  
x_whole = pd.concat([city_day_pred_whole, dt], axis=1)
x_whole
x_whole


#calculating the root mean squared error for the predicted AQI values
print(np.sqrt(metrics.mean_squared_error(y_whole,y_pred_whole)))

12.749887150652524


x_axis=x_whole.Year
y_axis=x_whole.Actual
y1_axis=x_whole.Predicted
plt.plot(x_axis,y_axis)
plt.plot(x_axis,y1_axis)
plt.title("Actual vs Predicted",fontsize=20)
plt.legend(["actual ","predicted"])
plt.xlabel("YEAR",fontsize=20)
plt.ylabel("AQI",fontsize=20)
plt.tick_params(labelsize=20)
plt.show()


city_day_pred_bfcovid = city_day_update_pred.copy()


city_day_pred_bfcovid = city_day_pred_bfcovid[city_day_pred_bfcovid['Date'] <= ('01-01-2020')]


city_day_pred_bfcovid


#Applying boxplot analysis
city_day_pred_bfcovid = city_day_pred_bfcovid[['AQI_avg','Year']].groupby(["Year"]).mean().reset_index().sort_values(by='Year',ascending=False)
city_day_pred_bfcovid = city_day_pred_bfcovid.dropna()
city_day_pred_bfcovid.describe()

sns.boxplot(x=city_day_pred_bfcovid['AQI_avg'])

<AxesSubplot:xlabel='AQI_avg'>


#Visualizing the filtered data

Year=city_day_pred_bfcovid['Year'].values
AQI=city_day_pred_bfcovid['AQI_avg'].values
city_day_pred_bfcovid['AQI_avg']=pd.to_numeric(city_day_pred_bfcovid['AQI_avg'],errors='coerce')
city_day_pred_bfcovid['Year']=pd.to_numeric(city_day_pred_bfcovid['Year'],errors='coerce')

#Plotting by using 3d graph
plt.rcParams['figure.figsize'] = (20.0, 10.0)
fig = plt.figure()
ax = fig.add_subplot(111, projection="3d")
ax.set_position([0, 0, 0.95, 1])
ax.scatter(Year,AQI, color='red')
plt.show()


#scatter plot of data points
cols =['Year']
y_bf = city_day_pred_bfcovid['AQI_avg']
x_bf = city_day_pred_bfcovid[cols]

plt.scatter(x_bf,y_bf)
plt.show()


x_bf = (x_bf - x_bf.mean()) / x_bf.std()
x_bf = np.c_[np.ones(x_bf.shape[0]), x_bf]
x_bf

array([[ 1.        ,  1.33630621],
       [ 1.        ,  0.80178373],
       [ 1.        ,  0.26726124],
       [ 1.        , -0.26726124],
       [ 1.        , -0.80178373],
       [ 1.        , -1.33630621]])


# Applying GRADIENT DESCENT 

alpha = 0.1 #Step size
iterations = 3000 #No. of iterations
m = y_bf.size #No. of data points
np.random.seed(4) #Setting the seed
theta = np.random.rand(2) #Picking random values to start with

def gradient_descent(x_bf, y_bf, theta, iterations, alpha):
    past_costs = []
    past_thetas = [theta]
    for i in range(iterations):
        prediction = np.dot(x_bf, theta)
        error = prediction - y_bf
        cost = 1/(2*m) * np.dot(error.T, error)
        past_costs.append(cost)
        theta = theta - (alpha * (1/m) * np.dot(x_bf.T, error))
        past_thetas.append(theta)
        
    return past_thetas, past_costs
past_thetas_bf, past_costs_bf = gradient_descent(x_bf, y_bf, theta, iterations, alpha)
theta_bf = past_thetas_bf[-1]

#Printing the results...
print("Gradient Descent: {:.2f}, {:.2f}".format(theta_bf[0], theta_bf[1]))

Gradient Descent: 177.87, 5.20


#Plotting the cost function...
plt.title('Cost Function J')
plt.xlabel('No. of iterations')
plt.ylabel('Cost')
plt.plot(past_costs_bf)
plt.show()


#Predicted value
newB=[ 177.87, 5.20]
def rmse(y_bf,y_pred_bf):
    rmse=np.sqrt(sum(y_bf-y_pred_bf))
    return rmse

   
y_pred_bf=x_bf.dot(newB)

dt = pd.DataFrame({'Actual': y_bf, 'Predicted': y_pred_bf})  
x_bf = pd.concat([city_day_pred_whole, dt], axis=1)
x_bf
x_bf


#calculating the root mean squared error for the predicted AQI values
print(np.sqrt(metrics.mean_squared_error(y_bf,y_pred_bf)))

16.55481596210976


x_axis=x_bf.Year
y_axis=x_bf.Actual
y1_axis=x_bf.Predicted
plt.plot(x_axis,y_axis)
plt.plot(x_axis,y1_axis)
plt.title("Actual vs Predicted",fontsize=20)
plt.legend(["actual ","predicted"])
plt.xlabel("YEAR",fontsize=20)
plt.ylabel("AQI",fontsize=20)
plt.tick_params(labelsize=20)
plt.show()


city_day_pred_afcovid = city_day_update_pred.copy()


city_day_pred_afcovid = city_day_pred_afcovid[city_day_pred_afcovid['Date'] >= ('01-01-2020')]


city_day_pred_afcovid


city_day_pred_afcovid.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 183 entries, 1826 to 2008
Data columns (total 4 columns):
 #   Column   Non-Null Count  Dtype         
---  ------   --------------  -----         
 0   Date     183 non-null    datetime64[ns]
 1   Month    183 non-null    int64         
 2   Year     183 non-null    int64         
 3   AQI_avg  183 non-null    float64       
dtypes: datetime64[ns](1), float64(1), int64(2)
memory usage: 7.1 KB


#Applying boxplot analysis
city_day_pred_afcovid = city_day_pred_afcovid[['AQI_avg','Month']].groupby(["Month"]).mean().reset_index().sort_values(by='Month',ascending=False)
city_day_pred_afcovid = city_day_pred_afcovid.dropna()
city_day_pred_afcovid.describe()

sns.boxplot(x=city_day_pred_afcovid['AQI_avg'])

<AxesSubplot:xlabel='AQI_avg'>


#Visualizing the filtered data

Month=city_day_pred_afcovid['Month'].values
AQI=city_day_pred_afcovid['AQI_avg'].values
city_day_pred_afcovid['AQI_avg']=pd.to_numeric(city_day_pred_afcovid['AQI_avg'],errors='coerce')
city_day_pred_afcovid['Month']=pd.to_numeric(city_day_pred_afcovid['Month'],errors='coerce')

#Plotting by using 3d graph
plt.rcParams['figure.figsize'] = (20.0, 10.0)
fig = plt.figure()
ax = fig.add_subplot(111, projection="3d")
ax.set_position([0, 0, 0.95, 1])
ax.scatter(Month,AQI, color='red')
plt.show()


#scatter plot of data points
cols =['Month']
y_af = city_day_pred_afcovid['AQI_avg']
x_af = city_day_pred_afcovid[cols]

plt.scatter(x_af,y_af)
plt.show()


x_af = (x_af - x_af.mean()) / x_af.std()
x_af = np.c_[np.ones(x_af.shape[0]), x_af]
x_af

array([[ 1.        ,  1.38873015],
       [ 1.        ,  0.9258201 ],
       [ 1.        ,  0.46291005],
       [ 1.        ,  0.        ],
       [ 1.        , -0.46291005],
       [ 1.        , -0.9258201 ],
       [ 1.        , -1.38873015]])


# Applying GRADIENT DESCENT 

alpha = 0.1 #Step size
iterations = 3000 #No. of iterations
m = y_af.size #No. of data points
np.random.seed(4) #Setting the seed
theta = np.random.rand(2) #Picking random values to start with

def gradient_descent(x_af, y_af, theta, iterations, alpha):
    past_costs = []
    past_thetas = [theta]
    for i in range(iterations):
        prediction = np.dot(x_af, theta)
        error = prediction - y_af
        cost = 1/(2*m) * np.dot(error.T, error)
        past_costs.append(cost)
        theta = theta - (alpha * (1/m) * np.dot(x_af.T, error))
        past_thetas.append(theta)
        
    return past_thetas, past_costs
past_thetas_af, past_costs_af = gradient_descent(x_af, y_af, theta, iterations, alpha)
theta_af = past_thetas_af[-1]

#Printing the results...
print("Gradient Descent: {:.2f}, {:.2f}".format(theta_af[0], theta_af[1]))

Gradient Descent: 111.53, -31.99


#Plotting the cost function...
plt.title('Cost Function J')
plt.xlabel('No. of iterations')
plt.ylabel('Cost')
plt.plot(past_costs_af)
plt.show()


#Predicted val
newB=[ 111.53, -31.99]
def rmse(y_af,y_pred_af):
    rmse=np.sqrt(sum(y_af-y_pred_af))
    return rmse

   
y_pred_af=x_af.dot(newB)

dt = pd.DataFrame({'Actual': y_af, 'Predicted': y_pred_af})  
x_af = pd.concat([city_day_pred_whole, dt], axis=1)
x_af
x_af


#calculating the root mean squared error for the predicted AQI values
print(np.sqrt(metrics.mean_squared_error(y_af,y_pred_af)))

15.84282666114951


x_axis=x_af.Year
y_axis=x_af.Actual
y1_axis=x_af.Predicted
plt.plot(x_axis,y_axis)
plt.plot(x_axis,y1_axis)
plt.title("Actual vs Predicted",fontsize=20)
plt.legend(["actual ","predicted"])
plt.xlabel("YEAR",fontsize=20)
plt.ylabel("AQI",fontsize=20)
plt.tick_params(labelsize=20)
plt.show()


city_day_pred = pd.read_csv("city_day_update.csv")


city_day_pred


# Finding unique cities available in dataset.
city_day_pred["City"].unique()

array(['Ahmedabad', 'Aizawl', 'Amaravati', 'Amritsar', 'Bengaluru',
       'Bhopal', 'Brajrajnagar', 'Chandigarh', 'Chennai', 'Coimbatore',
       'Delhi', 'Ernakulam', 'Gurugram', 'Guwahati', 'Hyderabad',
       'Jaipur', 'Jorapokhar', 'Kochi', 'Kolkata', 'Lucknow', 'Mumbai',
       'Patna', 'Shillong', 'Talcher', 'Thiruvananthapuram',
       'Visakhapatnam'], dtype=object)


# Finding count of unique rows of cities in the dataset.
city_day_pred["City"].value_counts()

Ahmedabad             2009
Delhi                 2009
Mumbai                2009
Bengaluru             2009
Lucknow               2009
Chennai               2009
Hyderabad             2006
Patna                 1858
Gurugram              1679
Visakhapatnam         1462
Amritsar              1221
Jorapokhar            1169
Jaipur                1114
Thiruvananthapuram    1112
Amaravati              951
Brajrajnagar           938
Talcher                925
Kolkata                814
Guwahati               502
Coimbatore             386
Shillong               310
Chandigarh             304
Bhopal                 289
Ernakulam              162
Kochi                  162
Aizawl                 113
Name: City, dtype: int64


delhi =  city_day_pred[city_day_pred['City'] == 'Delhi']


delhi


#Reducing the target dataset
delhi_aqi = delhi[['Date','AQI']]
delhi_aqi.reset_index(inplace = True,drop = True)


delhi_aqi


#Training the model over the training data set by using Prophet(time series forecasting library)
train_df = delhi_aqi
train_df.rename(mapper = {'Date':'ds','AQI':'y'},axis =1,inplace = True)
train_df

m = Prophet(holidays_prior_scale=0,seasonality_prior_scale=20,n_changepoints= 50,daily_seasonality=True)
m.fit(train_df)

01:18:43 - cmdstanpy - INFO - Chain [1] start processing
01:18:44 - cmdstanpy - INFO - Chain [1] done processing

<prophet.forecaster.Prophet at 0x1ee27236140>


#Creating a future dataset
future = m.make_future_dataframe(periods=365)
future.tail()


# Predicting the future behaviour of AQI in Delhi
forecast = m.predict(future)
forecast[['ds', 'yhat', 'yhat_lower', 'yhat_upper']].tail()


# Finding the Cross validation accuracy for getting the idea of our model how much it is accurate.
from prophet.diagnostics import mape,cross_validation,performance_metrics
df_cv = cross_validation(m, initial='1100 days', period='121 days', horizon = '365 days')
df_p = performance_metrics(df_cv)
print('Cross Validation accuracy:', (1 - df_p['mape'].mean())*100)

  0%|          | 0/5 [00:00<?, ?it/s]

01:18:56 - cmdstanpy - INFO - Chain [1] start processing
01:18:56 - cmdstanpy - INFO - Chain [1] done processing
01:19:06 - cmdstanpy - INFO - Chain [1] start processing
01:19:07 - cmdstanpy - INFO - Chain [1] done processing
01:19:16 - cmdstanpy - INFO - Chain [1] start processing
01:19:17 - cmdstanpy - INFO - Chain [1] done processing
01:19:25 - cmdstanpy - INFO - Chain [1] start processing
01:19:26 - cmdstanpy - INFO - Chain [1] done processing
01:19:36 - cmdstanpy - INFO - Chain [1] start processing
01:19:36 - cmdstanpy - INFO - Chain [1] done processing

Cross Validation accuracy: 67.97047760703944


# By this we get the upcoming trends for whole year.
# By this we can get an idea of monthly, daily and hourly behaviour also.
from prophet.plot import plot_plotly

fig = plot_plotly(m, forecast ,xlabel = 'Date',ylabel= 'AQI',figsize=(1000,750))

fig.show()


# Choosing only Patna city
patna =  city_day_pred[city_day_pred['City'] == 'Patna']


patna


#Reducing target dataset
patna_aqi = patna[['Date','AQI']]
patna_aqi.reset_index(inplace = True,drop = True)


#Training the model over the training data set by using Prophet(time series forecasting library)
train_df = patna_aqi
train_df.rename(mapper = {'Date':'ds','AQI':'y'},axis =1,inplace = True)
train_df

m = Prophet(holidays_prior_scale=0,seasonality_prior_scale=20,n_changepoints= 50,daily_seasonality=True)
m.fit(train_df)

01:19:47 - cmdstanpy - INFO - Chain [1] start processing
01:19:48 - cmdstanpy - INFO - Chain [1] done processing

<prophet.forecaster.Prophet at 0x1ee26e83cd0>


#Creating a future dataset.
future = m.make_future_dataframe(periods=365)
future.tail()


# Predicting the future behaviour of AQI in Patna.
forecast = m.predict(future)
forecast[['ds', 'yhat', 'yhat_lower', 'yhat_upper']].tail()


# Finding the Cross validation accuracy for getting the idea of our model how much it is accurate.
from prophet.diagnostics import mape,cross_validation,performance_metrics
df_cv = cross_validation(m, initial='1100 days', period='121 days', horizon = '365 days')
df_p = performance_metrics(df_cv)
print('Cross Validation accuracy:', (1 - df_p['mape'].mean())*100)

  0%|          | 0/4 [00:00<?, ?it/s]

01:19:59 - cmdstanpy - INFO - Chain [1] start processing
01:19:59 - cmdstanpy - INFO - Chain [1] done processing
01:20:09 - cmdstanpy - INFO - Chain [1] start processing
01:20:10 - cmdstanpy - INFO - Chain [1] done processing
01:20:17 - cmdstanpy - INFO - Chain [1] start processing
01:20:17 - cmdstanpy - INFO - Chain [1] done processing
01:20:27 - cmdstanpy - INFO - Chain [1] start processing
01:20:28 - cmdstanpy - INFO - Chain [1] done processing

Cross Validation accuracy: 63.38596674035115


# By this we get the upcoming trends for whole year.
# By this we can get an idea of monthly, daily and hourly behaviour also.
from prophet.plot import plot_plotly

fig = plot_plotly(m, forecast ,xlabel = 'Date',ylabel= 'AQI',figsize=(1000,750))

fig.show()


# Choosing only Bengaluru city
bengaluru =  city_day_pred[city_day_pred['City'] == 'Bengaluru']


bengaluru


#Reducing target data_set
bengaluru_aqi = bengaluru[['Date','AQI']]
bengaluru_aqi.reset_index(inplace = True,drop = True)


#Training the model over the training data set by using Prophet(time series forecasting library)
train_df = bengaluru_aqi
train_df.rename(mapper = {'Date':'ds','AQI':'y'},axis =1,inplace = True)
train_df

m = Prophet(holidays_prior_scale=0,seasonality_prior_scale=20,n_changepoints= 50,daily_seasonality=True)
m.fit(train_df)

01:20:40 - cmdstanpy - INFO - Chain [1] start processing
01:20:40 - cmdstanpy - INFO - Chain [1] done processing

<prophet.forecaster.Prophet at 0x1ee2cf6f580>


#Creating a future dataset.
future = m.make_future_dataframe(periods=365)
future.tail()


# Predicting the future behaviour of AQI in Bengaluru.
forecast = m.predict(future)
forecast[['ds', 'yhat', 'yhat_lower', 'yhat_upper']].tail()


# Finding the Cross validation accuracy for getting the idea of our model how much it is accurate.
from prophet.diagnostics import mape,cross_validation,performance_metrics
df_cv = cross_validation(m, initial='1100 days', period='121 days', horizon = '365 days')
df_p = performance_metrics(df_cv)
print('Cross Validation accuracy:', (1 - df_p['mape'].mean())*100)

  0%|          | 0/5 [00:00<?, ?it/s]

01:20:52 - cmdstanpy - INFO - Chain [1] start processing
01:20:53 - cmdstanpy - INFO - Chain [1] done processing
01:21:03 - cmdstanpy - INFO - Chain [1] start processing
01:21:03 - cmdstanpy - INFO - Chain [1] done processing
01:21:12 - cmdstanpy - INFO - Chain [1] start processing
01:21:13 - cmdstanpy - INFO - Chain [1] done processing
01:21:22 - cmdstanpy - INFO - Chain [1] start processing
01:21:23 - cmdstanpy - INFO - Chain [1] done processing
01:21:32 - cmdstanpy - INFO - Chain [1] start processing
01:21:33 - cmdstanpy - INFO - Chain [1] done processing

Cross Validation accuracy: 63.49849589422245


# By this we get the upcoming trends for whole year.
# By this we can get an idea of monthly, daily and hourly behaviour also.
from prophet.plot import plot_plotly

fig = plot_plotly(m, forecast ,xlabel = 'Date',ylabel= 'AQI',figsize=(1000,750))

fig.show()


# Choosing only Chennai city
chennai =  city_day_pred[city_day_pred['City'] == 'Chennai']


chennai


#Reducing target data_set
chennai_aqi = chennai[['Date','AQI']]
chennai_aqi.reset_index(inplace = True,drop = True)


#Training the model over the training data set by using Prophet(time series forecasting library)
train_df = chennai_aqi
train_df.rename(mapper = {'Date':'ds','AQI':'y'},axis =1,inplace = True)
train_df

m = Prophet(holidays_prior_scale=0,seasonality_prior_scale=20,n_changepoints= 50,daily_seasonality=True)
m.fit(train_df)

01:21:45 - cmdstanpy - INFO - Chain [1] start processing
01:21:46 - cmdstanpy - INFO - Chain [1] done processing

<prophet.forecaster.Prophet at 0x1ee27552b30>


#Creating a future dataset.
future = m.make_future_dataframe(periods=365)
future.tail()


# Predicting the future behaviour of AQI in Chennai.
forecast = m.predict(future)
forecast[['ds', 'yhat', 'yhat_lower', 'yhat_upper']].tail()


# Finding the Cross validation accuracy for getting the idea of our model how much it is accurate.
from prophet.diagnostics import mape,cross_validation,performance_metrics
df_cv = cross_validation(m, initial='1100 days', period='121 days', horizon = '365 days')
df_p = performance_metrics(df_cv)
print('Cross Validation accuracy:', (1 - df_p['mape'].mean())*100)

  0%|          | 0/5 [00:00<?, ?it/s]

01:21:57 - cmdstanpy - INFO - Chain [1] start processing
01:21:57 - cmdstanpy - INFO - Chain [1] done processing
01:22:07 - cmdstanpy - INFO - Chain [1] start processing
01:22:07 - cmdstanpy - INFO - Chain [1] done processing
01:22:17 - cmdstanpy - INFO - Chain [1] start processing
01:22:17 - cmdstanpy - INFO - Chain [1] done processing
01:22:27 - cmdstanpy - INFO - Chain [1] start processing
01:22:28 - cmdstanpy - INFO - Chain [1] done processing
01:22:37 - cmdstanpy - INFO - Chain [1] start processing
01:22:37 - cmdstanpy - INFO - Chain [1] done processing

Cross Validation accuracy: 61.76724564183594


# By this we get the upcoming trends for whole year.
# By this we can get an idea of monthly, daily and hourly behaviour also.
from prophet.plot import plot_plotly

fig = plot_plotly(m, forecast ,xlabel = 'Date',ylabel= 'AQI',figsize=(1000,750))

fig.show()

	StationId	StationName	City	State	Status
0	AP001	Secretariat, Amaravati - APPCB	Amaravati	Andhra Pradesh	Active
1	AP002	Anand Kala Kshetram, Rajamahendravaram - APPCB	Rajamahendravaram	Andhra Pradesh	NaN
2	AP003	Tirumala, Tirupati - APPCB	Tirupati	Andhra Pradesh	NaN
3	AP004	PWD Grounds, Vijayawada - APPCB	Vijayawada	Andhra Pradesh	NaN
4	AP005	GVM Corporation, Visakhapatnam - APPCB	Visakhapatnam	Andhra Pradesh	Active
...	...	...	...	...	...
225	WB010	Jadavpur, Kolkata - WBPCB	Kolkata	West Bengal	Active
226	WB011	Rabindra Bharati University, Kolkata - WBPCB	Kolkata	West Bengal	Active
227	WB012	Rabindra Sarobar, Kolkata - WBPCB	Kolkata	West Bengal	Active
228	WB013	Victoria, Kolkata - WBPCB	Kolkata	West Bengal	Active
229	WB014	Ward-32 Bapupara, Siliguri - WBPCB	Siliguri	West Bengal	NaN

	StationId	Date	PM2.5	PM10	NO	NO2	NOx	NH3	CO	SO2	O3	Benzene	Toluene	Xylene	AQI	AQI_Bucket
0	AP001	2017-11-24	71.36	115.75	1.75	20.65	12.40	12.19	0.10	10.76	109.26	0.17	5.92	0.10	NaN	NaN
1	AP001	2017-11-25	81.40	124.50	1.44	20.50	12.08	10.72	0.12	15.24	127.09	0.20	6.50	0.06	184.0	Moderate
2	AP001	2017-11-26	78.32	129.06	1.26	26.00	14.85	10.28	0.14	26.96	117.44	0.22	7.95	0.08	197.0	Moderate
3	AP001	2017-11-27	88.76	135.32	6.60	30.85	21.77	12.91	0.11	33.59	111.81	0.29	7.63	0.12	198.0	Moderate
4	AP001	2017-11-28	64.18	104.09	2.56	28.07	17.01	11.42	0.09	19.00	138.18	0.17	5.02	0.07	188.0	Moderate
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
108030	WB013	2020-06-27	8.65	16.46	NaN	NaN	NaN	NaN	0.69	4.36	30.59	1.32	7.26	NaN	50.0	Good
108031	WB013	2020-06-28	11.80	18.47	NaN	NaN	NaN	NaN	0.68	3.49	38.95	1.42	7.92	NaN	65.0	Satisfactory
108032	WB013	2020-06-29	18.60	32.26	13.65	200.87	214.20	11.40	0.78	5.12	38.17	3.52	8.64	NaN	63.0	Satisfactory
108033	WB013	2020-06-30	16.07	39.30	7.56	29.13	36.69	29.26	0.69	5.88	29.64	1.86	8.40	NaN	57.0	Satisfactory
108034	WB013	2020-07-01	10.50	36.50	7.78	22.50	30.25	27.23	0.58	2.80	13.10	1.31	7.39	NaN	59.0	Satisfactory

	StationId	StationName	City	State	Date	PM2.5	PM10	NO	NO2	NOx	NH3	CO	SO2	O3	Benzene	Toluene	Xylene	AQI	AQI_Bucket
0	AP001	Secretariat, Amaravati - APPCB	Amaravati	Andhra Pradesh	2017-11-24	71.36	115.75	1.75	20.65	12.40	12.19	0.10	10.76	109.26	0.17	5.92	0.10	NaN	NaN
1	AP001	Secretariat, Amaravati - APPCB	Amaravati	Andhra Pradesh	2017-11-25	81.40	124.50	1.44	20.50	12.08	10.72	0.12	15.24	127.09	0.20	6.50	0.06	184.0	Moderate
2	AP001	Secretariat, Amaravati - APPCB	Amaravati	Andhra Pradesh	2017-11-26	78.32	129.06	1.26	26.00	14.85	10.28	0.14	26.96	117.44	0.22	7.95	0.08	197.0	Moderate
3	AP001	Secretariat, Amaravati - APPCB	Amaravati	Andhra Pradesh	2017-11-27	88.76	135.32	6.60	30.85	21.77	12.91	0.11	33.59	111.81	0.29	7.63	0.12	198.0	Moderate
4	AP001	Secretariat, Amaravati - APPCB	Amaravati	Andhra Pradesh	2017-11-28	64.18	104.09	2.56	28.07	17.01	11.42	0.09	19.00	138.18	0.17	5.02	0.07	188.0	Moderate
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
107706	WB013	Victoria, Kolkata - WBPCB	Kolkata	West Bengal	2020-06-27	8.65	16.46	NaN	NaN	NaN	NaN	0.69	4.36	30.59	1.32	7.26	NaN	50.0	Good
107707	WB013	Victoria, Kolkata - WBPCB	Kolkata	West Bengal	2020-06-28	11.80	18.47	NaN	NaN	NaN	NaN	0.68	3.49	38.95	1.42	7.92	NaN	65.0	Satisfactory
107708	WB013	Victoria, Kolkata - WBPCB	Kolkata	West Bengal	2020-06-29	18.60	32.26	13.65	200.87	214.20	11.40	0.78	5.12	38.17	3.52	8.64	NaN	63.0	Satisfactory
107709	WB013	Victoria, Kolkata - WBPCB	Kolkata	West Bengal	2020-06-30	16.07	39.30	7.56	29.13	36.69	29.26	0.69	5.88	29.64	1.86	8.40	NaN	57.0	Satisfactory
107710	WB013	Victoria, Kolkata - WBPCB	Kolkata	West Bengal	2020-07-01	10.50	36.50	7.78	22.50	30.25	27.23	0.58	2.80	13.10	1.31	7.39	NaN	59.0	Satisfactory

	StationId	StationName	City	State	Date	PM2.5	PM10	NO	NO2	NOx	NH3	CO	SO2	O3	Benzene	Toluene	Xylene	AQI	Air_quality
0	AP001	Secretariat, Amaravati - APPCB	Amaravati	Andhra Pradesh	2017-11-24	71.36	115.75	1.75	20.65	12.40	12.19	0.10	10.76	109.26	0.17	5.92	0.10	184.0	Moderate
1	AP001	Secretariat, Amaravati - APPCB	Amaravati	Andhra Pradesh	2017-11-25	81.40	124.50	1.44	20.50	12.08	10.72	0.12	15.24	127.09	0.20	6.50	0.06	184.0	Moderate
2	AP001	Secretariat, Amaravati - APPCB	Amaravati	Andhra Pradesh	2017-11-26	78.32	129.06	1.26	26.00	14.85	10.28	0.14	26.96	117.44	0.22	7.95	0.08	197.0	Moderate
3	AP001	Secretariat, Amaravati - APPCB	Amaravati	Andhra Pradesh	2017-11-27	88.76	135.32	6.60	30.85	21.77	12.91	0.11	33.59	111.81	0.29	7.63	0.12	198.0	Moderate
4	AP001	Secretariat, Amaravati - APPCB	Amaravati	Andhra Pradesh	2017-11-28	64.18	104.09	2.56	28.07	17.01	11.42	0.09	19.00	138.18	0.17	5.02	0.07	188.0	Moderate
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
107706	WB013	Victoria, Kolkata - WBPCB	Kolkata	West Bengal	2020-06-27	8.65	16.46	NaN	NaN	NaN	NaN	0.69	4.36	30.59	1.32	7.26	NaN	50.0	Good
107707	WB013	Victoria, Kolkata - WBPCB	Kolkata	West Bengal	2020-06-28	11.80	18.47	13.65	200.87	214.20	11.40	0.68	3.49	38.95	1.42	7.92	NaN	65.0	Satisfactory
107708	WB013	Victoria, Kolkata - WBPCB	Kolkata	West Bengal	2020-06-29	18.60	32.26	13.65	200.87	214.20	11.40	0.78	5.12	38.17	3.52	8.64	NaN	63.0	Satisfactory
107709	WB013	Victoria, Kolkata - WBPCB	Kolkata	West Bengal	2020-06-30	16.07	39.30	7.56	29.13	36.69	29.26	0.69	5.88	29.64	1.86	8.40	NaN	57.0	Satisfactory
107710	WB013	Victoria, Kolkata - WBPCB	Kolkata	West Bengal	2020-07-01	10.50	36.50	7.78	22.50	30.25	27.23	0.58	2.80	13.10	1.31	7.39	NaN	59.0	Satisfactory

	City	Date	PM2.5	PM10	NO	NO2	NOx	NH3	CO	SO2	O3	Benzene	Toluene	Xylene	AQI	AQI_Bucket
0	Ahmedabad	2015-01-01	NaN	NaN	0.92	18.22	17.15	NaN	0.92	27.64	133.36	0.00	0.02	0.00	NaN	NaN
1	Ahmedabad	2015-01-02	NaN	NaN	0.97	15.69	16.46	NaN	0.97	24.55	34.06	3.68	5.50	3.77	NaN	NaN
2	Ahmedabad	2015-01-03	NaN	NaN	17.40	19.30	29.70	NaN	17.40	29.07	30.70	6.80	16.40	2.25	NaN	NaN
3	Ahmedabad	2015-01-04	NaN	NaN	1.70	18.48	17.97	NaN	1.70	18.59	36.08	4.43	10.14	1.00	NaN	NaN
4	Ahmedabad	2015-01-05	NaN	NaN	22.10	21.42	37.76	NaN	22.10	39.33	39.31	7.01	18.89	2.78	NaN	NaN
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
29526	Visakhapatnam	2020-06-27	15.02	50.94	7.68	25.06	19.54	12.47	0.47	8.55	23.30	2.24	12.07	0.73	41.0	Good
29527	Visakhapatnam	2020-06-28	24.38	74.09	3.42	26.06	16.53	11.99	0.52	12.72	30.14	0.74	2.21	0.38	70.0	Satisfactory
29528	Visakhapatnam	2020-06-29	22.91	65.73	3.45	29.53	18.33	10.71	0.48	8.42	30.96	0.01	0.01	0.00	68.0	Satisfactory
29529	Visakhapatnam	2020-06-30	16.64	49.97	4.05	29.26	18.80	10.03	0.52	9.84	28.30	0.00	0.00	0.00	54.0	Satisfactory
29530	Visakhapatnam	2020-07-01	15.00	66.00	0.40	26.85	14.05	5.20	0.59	2.10	17.05	NaN	NaN	NaN	50.0	Good

	Missing Values	% of Total Values
Xylene	84595	78.500000
NH3	47245	43.900000
PM10	41789	38.800000
Toluene	37453	34.800000
Benzene	30164	28.000000
O3	24213	22.500000
SO2	23922	22.200000
PM2.5	20417	19.000000
AQI	18958	17.600000
Air_quality	18958	17.600000
NO	15629	14.500000
NO2	15058	14.000000
NOx	14346	13.300000
CO	11386	10.600000

	Missing Values	% of Total Values
Xylene	17878	60.500000
PM10	10866	36.800000
NH3	10061	34.100000
Toluene	7739	26.200000
Benzene	5298	17.900000
PM2.5	4321	14.600000
AQI	4174	14.100000
Air_quality	4174	14.100000
NOx	3980	13.500000
O3	3664	12.400000
SO2	3510	11.900000
NO2	3278	11.100000
NO	3276	11.100000
CO	1745	5.900000

	StationName	PM2.5
0	Anand Vihar, Delhi - DPCC	152.350000
1	Talkatora District Industries Center, Lucknow - CPCB	134.690000
2	DTU, Delhi - CPCB	131.080000
3	IGSC Planetarium Complex, Patna - BSPCB	130.450000
4	Jahangirpuri, Delhi - DPCC	128.120000
5	Wazirpur, Delhi - DPCC	127.510000
6	Mundka, Delhi - DPCC	122.460000
7	Rohini, Delhi - DPCC	122.420000
8	Bawana, Delhi - DPCC	120.950000
9	Burari Crossing, Delhi - IMD	120.820000

	StationName	PM10
0	Anand Vihar, Delhi - DPCC	358.120000
1	Wazirpur, Delhi - DPCC	277.420000
2	Dwarka-Sector 8, Delhi - DPCC	276.470000
3	Mundka, Delhi - DPCC	269.130000
4	Jahangirpuri, Delhi - DPCC	259.450000
5	Sirifort, Delhi - CPCB	252.480000
6	Rohini, Delhi - DPCC	247.870000
7	NSIT Dwarka, Delhi - CPCB	242.770000
8	R K Puram, Delhi - DPCC	242.410000
9	DTU, Delhi - CPCB	236.060000

	StationName	Particulate_Matter
0	Anand Vihar, Delhi - DPCC	509.740000
1	Wazirpur, Delhi - DPCC	405.510000
2	Mundka, Delhi - DPCC	391.590000
3	Jahangirpuri, Delhi - DPCC	387.750000
4	Dwarka-Sector 8, Delhi - DPCC	379.790000
5	Rohini, Delhi - DPCC	370.440000
6	R K Puram, Delhi - DPCC	361.170000
7	Bawana, Delhi - DPCC	354.700000
8	Sirifort, Delhi - CPCB	349.750000
9	DTU, Delhi - CPCB	346.270000

	StationName	NO
0	Samanpura, Patna - BSPCB	124.000000
1	Anand Vihar, Delhi - DPCC	90.860000
2	Pusa, Delhi - DPCC	73.060000
3	DRM Office Danapur, Patna - BSPCB	64.930000
4	Major Dhyan Chand National Stadium, Delhi - DPCC	57.780000
5	R K Puram, Delhi - DPCC	54.420000
6	Chhatrapati Shivaji Intl. Airport (T2), Mumbai - MPCB	53.380000
7	Jawaharlal Nehru Stadium, Delhi - DPCC	52.570000
8	ITO, Delhi - CPCB	50.600000
9	Sirifort, Delhi - CPCB	46.990000

	StationName	NO2
0	Anand Vihar, Delhi - DPCC	88.720000
1	Punjabi Bagh, Delhi - DPCC	73.280000
2	Rajbansi Nagar, Patna - BSPCB	65.900000
3	Jahangirpuri, Delhi - DPCC	65.860000
4	Jawaharlal Nehru Stadium, Delhi - DPCC	63.340000
5	R K Puram, Delhi - DPCC	63.010000
6	Pusa, Delhi - DPCC	59.570000
7	Sirifort, Delhi - CPCB	58.860000
8	Maninagar, Ahmedabad - GPCB	58.850000
9	Major Dhyan Chand National Stadium, Delhi - DPCC	58.720000

	StationName	NOx
0	Anand Vihar, Delhi - DPCC	148.780000
1	Samanpura, Patna - BSPCB	141.460000
2	East Arjun Nagar, Delhi - CPCB	121.800000
3	Pusa, Delhi - DPCC	91.230000
4	R K Puram, Delhi - DPCC	86.360000
5	Chhatrapati Shivaji Intl. Airport (T2), Mumbai - MPCB	84.670000
6	Major Dhyan Chand National Stadium, Delhi - DPCC	82.100000
7	Jawaharlal Nehru Stadium, Delhi - DPCC	81.310000
8	Sion, Mumbai - MPCB	74.310000
9	Victoria, Kolkata - WBPCB	73.730000

	StationName	NH3
0	Manali, Chennai - CPCB	65.360000
1	Anand Vihar, Delhi - DPCC	55.780000
2	Jahangirpuri, Delhi - DPCC	55.670000
3	Rohini, Delhi - DPCC	53.010000
4	ITO, Delhi - CPCB	52.060000
5	IGSC Planetarium Complex, Patna - BSPCB	51.240000
6	NSIT Dwarka, Delhi - CPCB	48.870000
7	Patparganj, Delhi - DPCC	48.850000
8	Shadipur, Delhi - CPCB	45.680000
9	Mundka, Delhi - DPCC	45.510000

	StationName	CO
0	Maninagar, Ahmedabad - GPCB	22.360000
1	BWSSB Kadabesanahalli, Bengaluru - CPCB	3.580000
2	Shadipur, Delhi - CPCB	3.480000
3	Peenya, Bengaluru - CPCB	3.010000
4	NSIT Dwarka, Delhi - CPCB	2.840000
5	Central School, Lucknow - CPCB	2.310000
6	Lalbagh, Lucknow - CPCB	2.260000
7	Anand Vihar, Delhi - DPCC	2.200000
8	ITO, Delhi - CPCB	2.120000
9	Alandur Bus Depot, Chennai - CPCB	1.950000

	StationName	SO2
0	Maninagar, Ahmedabad - GPCB	55.250000
1	Tata Stadium, Jorapokhar - JSPCB	34.640000
2	Talcher Coalfields,Talcher - OSPCB	28.410000
3	Pusa, Delhi - IMD	27.630000
4	Lodhi Road, Delhi - IMD	23.250000
5	North Campus, DU, Delhi - IMD	23.250000
6	IGSC Planetarium Complex, Patna - BSPCB	22.980000
7	R K Puram, Delhi - DPCC	20.690000
8	Punjabi Bagh, Delhi - DPCC	20.080000
9	Alipur, Delhi - DPCC	19.860000

	StationName	O3
0	Punjabi Bagh, Delhi - DPCC	230.100000
1	Sector-51, Gurugram - HSPCB	80.750000
2	Manali Village, Chennai - TNPCB	69.610000
3	T T Nagar, Bhopal - MPPCB	59.940000
4	R K Puram, Delhi - DPCC	54.780000
5	Shastri Nagar, Jaipur - RSPCB	53.200000
6	Teri Gram, Gurugram - HSPCB	52.960000
7	Adarsh Nagar, Jaipur - RSPCB	51.530000
8	Sirifort, Delhi - CPCB	49.010000
9	Hombegowda Nagar, Bengaluru - KSPCB	47.610000

	StationName	BTX
0	Jadavpur, Kolkata - WBPCB	220.430000
1	Talkatora District Industries Center, Lucknow - CPCB	56.030000
2	Maninagar, Ahmedabad - GPCB	37.630000
3	Burari Crossing, Delhi - IMD	33.190000
4	IDA Pashamylaram, Hyderabad - TSPCB	31.310000
5	Fort William, Kolkata - WBPCB	29.760000
6	Bidhannagar, Kolkata - WBPCB	27.680000
7	Ballygunge, Kolkata - WBPCB	26.130000
8	Mandir Marg, Delhi - DPCC	20.830000
9	Teri Gram, Gurugram - HSPCB	20.500000

	StationName	PM2.5
0	City Railway Station, Bengaluru - KSPCB	9.000000
1	East Arjun Nagar, Delhi - CPCB	11.110000
2	Sikulpuikawn, Aizawl - Mizoram PCB	16.850000
3	Manali Village, Chennai - TNPCB	24.480000
4	Plammoodu, Thiruvananthapuram - Kerala PCB	27.220000
5	Hombegowda Nagar, Bengaluru - KSPCB	27.520000
6	Hebbal, Bengaluru - KSPCB	28.930000
7	Kariavattom, Thiruvananthapuram - Kerala PCB	28.980000
8	SIDCO Kurichi, Coimbatore - TNPCB	29.090000
9	Borivali East, Mumbai - MPCB	29.290000

	StationName	PM10
0	East Arjun Nagar, Delhi - CPCB	6.320000
1	Sikulpuikawn, Aizawl - Mizoram PCB	23.340000
2	Talkatora District Industries Center, Lucknow - CPCB	26.860000
3	SIDCO Kurichi, Coimbatore - TNPCB	37.740000
4	BWSSB Kadabesanahalli, Bengaluru - CPCB	40.750000
5	Lumpyngngad, Shillong - Meghalaya PCB	41.640000
6	Velachery Res. Area, Chennai - CPCB	43.490000
7	Alandur Bus Depot, Chennai - CPCB	48.550000
8	Kariavattom, Thiruvananthapuram - Kerala PCB	51.490000
9	Plammoodu, Thiruvananthapuram - Kerala PCB	52.190000

	StationName	Particulate_Matter
0	East Arjun Nagar, Delhi - CPCB	17.430000
1	City Railway Station, Bengaluru - KSPCB	40.000000
2	Sikulpuikawn, Aizawl - Mizoram PCB	40.190000
3	Sanegurava Halli, Bengaluru - KSPCB	45.960000
4	Alandur Bus Depot, Chennai - CPCB	56.320000
5	Talkatora District Industries Center, Lucknow - CPCB	57.880000
6	Velachery Res. Area, Chennai - CPCB	60.770000
7	BWSSB Kadabesanahalli, Bengaluru - CPCB	62.450000
8	SIDCO Kurichi, Coimbatore - TNPCB	66.900000
9	Lumpyngngad, Shillong - Meghalaya PCB	72.880000

	StationName	NO
0	Lumpyngngad, Shillong - Meghalaya PCB	0.920000
1	Bollaram Industrial Area, Hyderabad - TSPCB	2.960000
2	Borivali East, Mumbai - MPCB	3.260000
3	Plammoodu, Thiruvananthapuram - Kerala PCB	3.410000
4	ICRISAT Patancheru, Hyderabad - TSPCB	3.550000
5	Hombegowda Nagar, Bengaluru - KSPCB	3.570000
6	Sector-51, Gurugram - HSPCB	3.810000
7	Secretariat, Amaravati - APPCB	4.450000
8	Peenya, Bengaluru - CPCB	4.510000
9	Powai, Mumbai - MPCB	4.770000

	StationName	NO2
0	Sikulpuikawn, Aizawl - Mizoram PCB	0.390000
1	Lumpyngngad, Shillong - Meghalaya PCB	2.770000
2	Borivali East, Mumbai - MPCB	4.650000
3	Teri Gram, Gurugram - HSPCB	4.910000
4	Manali Village, Chennai - TNPCB	9.110000
5	Plammoodu, Thiruvananthapuram - Kerala PCB	9.190000
6	Tata Stadium, Jorapokhar - JSPCB	9.370000
7	Govt. High School Shikarpur, Patna - BSPCB	9.620000
8	Powai, Mumbai - MPCB	10.390000
9	Sector-25, Chandigarh - CPCC	11.630000

	StationName	NOx
0	Lumpyngngad, Shillong - Meghalaya PCB	1.000000
1	Teri Gram, Gurugram - HSPCB	5.960000
2	Tata Stadium, Jorapokhar - JSPCB	7.410000
3	Plammoodu, Thiruvananthapuram - Kerala PCB	7.550000
4	Borivali East, Mumbai - MPCB	7.710000
5	Govt. High School Shikarpur, Patna - BSPCB	9.440000
6	Sector-51, Gurugram - HSPCB	11.210000
7	ICRISAT Patancheru, Hyderabad - TSPCB	11.220000
8	Bollaram Industrial Area, Hyderabad - TSPCB	12.140000
9	Sikulpuikawn, Aizawl - Mizoram PCB	12.610000

	StationName	NH3
0	Lumpyngngad, Shillong - Meghalaya PCB	2.810000
1	Plammoodu, Thiruvananthapuram - Kerala PCB	5.030000
2	Worli, Mumbai - MPCB	6.560000
3	Tata Stadium, Jorapokhar - JSPCB	7.000000
4	Bandra, Mumbai - MPCB	7.160000
5	Colaba, Mumbai - MPCB	8.020000
6	Kariavattom, Thiruvananthapuram - Kerala PCB	8.040000
7	Borivali East, Mumbai - MPCB	8.170000
8	Nishant Ganj, Lucknow - UPPCB	8.910000
9	SIDCO Kurichi, Coimbatore - TNPCB	9.310000

Air Quality Measure based on AQI¶

About The Project¶

Libraries¶

Data Collection¶

Data Preprocessing and Cleaning¶

Cleaning of Stations' Data to remove the missing values¶

Merging two dataframes i.e. station_day and stations on basis of StationId.¶

Cleaning of Cities' Data to remove the missing values¶

Data Visualisation¶

Descriptive Analysis of Stations_day¶

Calculating the number of null values in each columns.¶

Descriptive Analysis of city_day¶

Calculating the number of null values in each columns.¶

Data Statistics¶

Most and Least Polluted Stations based on the given dataset¶

Most Polluted¶

Least Polluted¶

Most and Least Polluted Cities based on the given dataset¶

Most Polluted¶

Least Polluted¶

Data Interpretation¶

Stations¶

Replacing zeroes with median values.¶

Cities¶

Replacing zeroes with median values.¶

Checking the data via various method for visualisation.¶

Hypothesis Testing¶

Visualisation of Air Quality before Covid and After Covid.¶

Before Covid¶

Plotting some visual data for further interpretation¶

Plotting the Most Polluted cities in the Bar form on the basis of types of pollutants and total pollutants¶

Plotting the Least Polluted cities in the Bar form on the basis of types of pollutants and total pollutants¶

Satisfaction Level of people in the cities that are majorly polluted¶

Before doing the hypothesis testing we are doing the normalisation and standardisation of dataset¶

Why is Normalization is important?¶

After Covid¶

Plotting some visual data for further interpretation**¶

Plotting the Most Polluted cities in the Bar form on the basis of types of pollutants and total pollutants¶

Plotting the Least Polluted cities in the Bar form on the basis of types of pollutants and total pollutants¶

Satisfaction Level of people in the cities that are majorly polluted¶

We're using the Standard Scaling method in which Mean is set to 0 and Variance to 1.¶

Predicting the AQI¶

1. Calculating the AQI as whole of India yearly by using Linear Regression¶

Predicting the AQI for taking all years in consideration¶

Predicting the AQI by taking the years before COVID-19¶

Predicting the AQI by taking the years after COVID-19¶

2. Calculating the AQI for each cities for upcoming years¶

I) Forecasting Delhi's AQI for the upcoming years by using the dataset available¶

II) Forecasting Patna's AQI for the upcoming years by using the dataset available¶

III) Forecasting Bengaluru's AQI for the upcoming years by using the dataset available¶

IV) Forecasting Chennai's AQI for the upcoming years by using the dataset available¶

From the above examples of 4 cities, by using this model, we are getting a cross-validation accuracy above 60. This shows that we can implement this model for predicting or forecasting the behaviour of AQI in future for various cities across India.¶

	StationName	CO
0	Lumpyngngad, Shillong - Meghalaya PCB	0.240000
1	Sikulpuikawn, Aizawl - Mizoram PCB	0.280000
2	Borivali East, Mumbai - MPCB	0.370000
3	Bollaram Industrial Area, Hyderabad - TSPCB	0.410000
4	Worli, Mumbai - MPCB	0.410000
5	Kurla, Mumbai - MPCB	0.420000
6	Colaba, Mumbai - MPCB	0.460000
7	Sanegurava Halli, Bengaluru - KSPCB	0.480000
8	ICRISAT Patancheru, Hyderabad - TSPCB	0.490000
9	Sion, Mumbai - MPCB	0.490000

	StationName	SO2
0	Kariavattom, Thiruvananthapuram - Kerala PCB	3.230000
1	BWSSB Kadabesanahalli, Bengaluru - CPCB	3.810000
2	Sanegurava Halli, Bengaluru - KSPCB	3.880000
3	DRM Office Danapur, Patna - BSPCB	4.780000
4	Silk Board, Bengaluru - KSPCB	4.820000
5	Zoo Park, Hyderabad - TSPCB	4.830000
6	Patparganj, Delhi - DPCC	4.840000
7	Central University, Hyderabad - TSPCB	5.150000
8	Jayanagar 5th Block, Bengaluru - KSPCB	5.270000
9	Muradpur, Patna - BSPCB	5.310000

	StationName	BTX
0	T T Nagar, Bhopal - MPPCB	0.000000
1	Bandra, Mumbai - MPCB	0.030000
2	SIDCO Kurichi, Coimbatore - TNPCB	0.200000
3	Lodhi Road, Delhi - IMD	1.960000
4	North Campus, DU, Delhi - IMD	2.380000
5	CRRI Mathura Road, Delhi - IMD	2.770000
6	Bollaram Industrial Area, Hyderabad - TSPCB	2.960000
7	ICRISAT Patancheru, Hyderabad - TSPCB	3.440000
8	Pusa, Delhi - IMD	3.710000
9	Secretariat, Amaravati - APPCB	3.860000

	City	PM2.5
0	Patna	123.110000
1	Gurugram	117.340000
2	Delhi	117.150000
3	Lucknow	109.940000
4	Ahmedabad	67.820000
5	Jorapokhar	64.670000
6	Brajrajnagar	64.360000
7	Kolkata	64.120000
8	Guwahati	63.940000
9	Talcher	61.010000

	City	PM10
0	Delhi	232.730000
1	Gurugram	192.490000
2	Talcher	165.290000
3	Jorapokhar	150.390000
4	Patna	126.910000
5	Brajrajnagar	124.940000
6	Jaipur	123.400000
7	Bhopal	119.210000
8	Guwahati	116.600000
9	Kolkata	115.260000

	City	NO
0	Kochi	71.370000
1	Delhi	38.980000
2	Patna	31.800000
3	Talcher	31.770000
4	Mumbai	31.560000
5	Kolkata	26.840000
6	Ernakulam	23.570000
7	Ahmedabad	22.590000
8	Guwahati	20.010000
9	Brajrajnagar	19.200000

	City	NOx
0	Jorapokhar	99.990000
1	Kochi	68.410000
2	Kolkata	63.340000
3	Delhi	58.570000
4	Mumbai	55.180000
5	Ahmedabad	47.370000
6	Patna	46.110000
7	Guwahati	44.250000
8	Jaipur	39.650000
9	Amritsar	35.690000

	City	NH3
0	Chennai	63.400000
1	Delhi	41.990000
2	Brajrajnagar	36.960000
3	Chandigarh	30.600000
4	Lucknow	29.220000
5	Ahmedabad	26.640000
6	Jaipur	26.470000
7	Gurugram	26.210000
8	Aizawl	22.310000
9	Bengaluru	22.160000

	City	BTX
0	Kolkata	38.110000
1	Ahmedabad	37.630000
2	Delhi	26.780000
3	Thiruvananthapuram	22.350000
4	Patna	17.270000
5	Visakhapatnam	15.080000
6	Gurugram	14.640000
7	Amritsar	14.500000
8	Hyderabad	10.720000
9	Lucknow	10.410000

	City	PM10
0	Aizawl	23.340000
1	Coimbatore	39.230000
2	Shillong	41.640000
3	Ernakulam	48.310000
4	Thiruvananthapuram	52.790000
5	Chennai	62.950000
6	Kochi	67.340000
7	Amaravati	76.310000
8	Bengaluru	83.590000
9	Chandigarh	85.660000